A Mixture of Urinary Phthalate Metabolite Concentrations During Pregnancy and Offspring Social Responsiveness Scale Scores.

BACKGROUND: Phthalates are a group of chemicals with ubiquitous exposure worldwide. Exposures to phthalates during pregnancy may play a role in autism spectrum disorder (ASD) etiology by disrupting hormone levels or directly impacting fetal neurodevelopment. However, there is little research quantifying the aggregate effect of phthalates on child ASD-related behaviors. METHODS: We used data from two prospective pregnancy and birth cohorts-the Health Outcomes and Measures of the Environment (HOME) and the Early Autism Risk Longitudinal Investigation (EARLI). HOME is a general population cohort while participants in EARLI were at higher familial risk for ASD. Using quantile g-computation and linear regression models, we assessed the joint and individual associations of a mixture of six phthalate metabolites during pregnancy with child ASD-related traits measured by Social Responsiveness Scale (SRS) scores at ages 3-8 years. RESULTS: Our analyses included 271 participants from HOME and 166 participants from EARLI. There were imprecise associations between the phthalate mixture and SRS total raw scores in HOME (difference in SRS scores per decile increase in every phthalate = 1.3; 95% confidence interval [CI] = -0.2, 2.8) and EARLI (difference in SRS scores per decile increase in every phthalate = -0.9; 95% CI = -3.5, 1.7). CONCLUSIONS: The cohort-specific effect sizes of the pthalates-SRS associations were small and CIs were imprecise. These results suggest that if there are associations between phthalate metabolites during pregnancy and child SRS scores, they may differ across populations with different familial liabilities. Further studies with larger sample sizes are warranted.

The Environmental Influences on Child Health Outcomes (ECHO)-Wide Cohort.

The Environmental Influences on Child Health Outcomes (ECHO)-Wide Cohort Study (EWC), a collaborative research design comprising 69 cohorts in 31 consortia, was funded by the National Institutes of Health (NIH) in 2016 to improve children's health in the United States. The EWC harmonizes extant data and collects new data using a standardized protocol, the ECHO-Wide Cohort Data Collection Protocol (EWCP). EWCP visits occur at least once per life stage, but the frequency and timing of the visits vary across cohorts. As of March 4, 2022, the EWC cohorts contributed data from 60,553 children and consented 29,622 children for new EWCP data and biospecimen collection. The median (interquartile range) age of EWCP-enrolled children was 7.5 years (3.7-11.1). Surveys, interviews, standardized examinations, laboratory analyses, and medical record abstraction are used to obtain information in 5 main outcome areas: pre-, peri-, and postnatal outcomes; neurodevelopment; obesity; airways; and positive health. Exposures include factors at the level of place (e.g., air pollution, neighborhood socioeconomic status), family (e.g., parental mental health), and individuals (e.g., diet, genomics).

Prenatal Exposure to Per- and Polyfluoroalkyl Substances and Childhood Autism-related Outcomes.

BACKGROUND: Epidemiologic evidence linking prenatal exposure to per- and polyfluoroalkyl substances (PFAS) with altered neurodevelopment is inconclusive, and few large studies have focused on autism-related outcomes. We investigated whether blood concentrations of PFAS in pregnancy are associated with child autism-related outcomes. METHODS: We included 10 cohorts from the National Institutes of Health (NIH)-funded Environmental influences on Child Health Outcomes (ECHO) program (n = 1,429). We measured 14 PFAS analytes in maternal blood collected during pregnancy; eight analytes met detection criteria for analysis. We assessed quantitative autism-related traits in children via parent report on the Social Responsiveness Scale (SRS). In multivariable linear models, we examined relationships of each PFAS (natural log-transformed) with SRS scores. We further modeled PFAS as a complex mixture using Bayesian methods and examined modification of these relationships by child sex. RESULTS: Most PFAS in maternal blood were not associated with child SRS T-scores. Perfluorononanoic acid (PFNA) showed the strongest and most consistent association: each 1-unit increase in ln-transformed PFNA was associated with greater autism-related traits (adjusted ß [95% confidence interval (CI)] = 1.5 [-0.1, 3.0]). The summed mixture, which included six PFAS detected in >70% of participants, was not associated with SRS T-scores (adjusted ß [95% highest posterior density interval] = 0.7 [-1.4, 3.0]). We did not observe consistent evidence of sex differences. CONCLUSIONS: Prenatal blood concentrations of PFNA may be associated with modest increases in child autism-related traits. Future work should continue to examine the relationship between exposures to both legacy and emerging PFAS and additional dimensional, quantitative measures of childhood autism-related outcomes.

Associations of prenatal exposure to a mixture of persistent organic pollutants with social traits and cognitive and adaptive function in early childhood: Findings from the EARLI study.

BACKGROUND: Literature suggests that maternal exposure to persistent organic pollutants (POPs) may influence child neurodevelopment. Evidence linking prenatal POPs and autism spectrum disorder has been inconclusive and few studies have examined the mixture effect of the POPs on autism-related traits. OBJECTIVE: To evaluate the associations between prenatal exposure to a mixture of POPs and autism-related traits in children from the Early Autism Risk Longitudinal Investigation study. METHODS: Maternal serum concentrations of 17 POPs (11 polychlorinated biphenyls [PCBs], 4 polybrominated diphenyls [PBDEs], and 2 persistent pesticides) in 154 samples collected during pregnancy were included in this analysis. We examined the independent associations of the natural log-transformed POPs with social, cognitive, and behavioral traits at 36 months of age, including Social Responsiveness Scale (SRS), Mullen Scales of Early Learning-Early Learning Composite (MSEL-ELC), and Vineland Adaptive Behavior Scales (VABS) scores, using linear regression models. We applied Bayesian kernel machine regression and quantile g-computation to examine the joint effect and interactions of the POPs. RESULTS: Higher ln-PBDE47 was associated with greater deficits in social reciprocity (higher SRS score) (ß = 6.39, 95% CI: 1.12, 11.65) whereas higher ln-p,p'-DDE was associated with lower social deficits (ß = -8.34, 95% CI: -15.32, -1.37). Positive associations were observed between PCB180 and PCB187 and cognitive (MSEL-ELC) scores (ß = 5.68, 95% CI: 0.18, 11.17; ß = 4.65, 95% CI: 0.14, 9.17, respectively). Adaptive functioning (VABS) scores were positively associated with PCB170, PCB180, PCB187, PCB196/203, and p,p'-DDE. In the mixture analyses, we did not observe an overall mixture effect of POPs on the quantitative traits. Potential interactions between PBDE99 and other PBDEs were identified in association with MSEL-ELC scores. CONCLUSIONS: We observed independent effects of PCB180, PCB187, PBDE47, and p,p' DDE with ASD-related quantitative traits and potential interactions between PBDEs. Our findings highlight the importance of assessing the effect of POPs as a mixture.

Invited Commentary: Is Bias Towards the Null From Nondifferential Misclassification Wishful Thinking?

In their article, Yland et al. (Am J Epidemiol. 2022;191(8):1485-1495) bring renewed attention to the oft-cited heuristic "nondifferential misclassification biases results towards the null." They make a compelling case, as others have before, that this heuristic is misguided. Herein, I briefly discuss the history of this heuristic, discuss why I believe it became and stayed popular, and note the very limited circumstances for which it may be appropriate.

Combining Effect Estimates Across Cohorts and Sufficient Adjustment Sets for Collaborative Research: A Simulation Study.

BACKGROUND: Collaborative research often combines findings across multiple, independent studies via meta-analysis. Ideally, all study estimates that contribute to the meta-analysis will be equally unbiased. Many meta-analyses require all studies to measure the same covariates. We explored whether differing minimally sufficient sets of confounders identified by a directed acyclic graph (DAG) ensures comparability of individual study estimates. Our analysis applied four statistical estimators to multiple minimally sufficient adjustment sets identified in a single DAG. METHODS: We compared estimates obtained via linear, log-binomial, and logistic regression and inverse probability weighting, and data were simulated based on a previously published DAG. RESULTS: Our results show that linear, log-binomial, and inverse probability weighting estimators generally provide the same estimate of effect for different estimands that are equally sufficient to adjust confounding bias, with modest differences in random error. In contrast, logistic regression often performed poorly, with notable differences in effect estimates obtained from unique minimally sufficient adjustment sets, and larger standard errors than other estimators. CONCLUSIONS: Our findings do not support the reliance of collaborative research on logistic regression results for meta-analyses. Use of DAGs to identify potentially differing minimally sufficient adjustment sets can allow meta-analyses without requiring the exact same covariates.

Motor Vehicle Crashes Involving a Bicycle Before and After Introduction of a Bike Share Program in Philadelphia, Pennsylvania, 2010-2018.

Objectives. To quantify the impact of a citywide bicycle share program on rates of motor vehicle collisions involving a bicycle.Methods. We conducted an interrupted time series analysis, using crash records from the Pennsylvania Department of Transportation for Philadelphia County from 2010 through 2018. We also calculated summary statistics to illustrate annual and monthly trends in rates of motor vehicle crashes involving a bicycle.Results. The baseline rate of bike events was 106% greater (95% confidence interval [CI] = 1.25, 3.38) at the time bicycle share was implemented compared with January 2010. Before bicycle share implementation, the rate of bicycle events decreased 1% (95% CI = 0.95, 1.03) annually. After the bicycle share program started, the rate of bicycle events decreased 13% (95% CI = 0.82, 0.94) annually.Conclusions. In the long term, programs that increase the number of bicycles on the road, such as bike share, may reduce rates of motor vehicle crashes involving a bicycle.

Prenatal Exposure to Endocrine-disrupting Chemicals in Relation to Autism Spectrum Disorder and Intellectual Disability.

BACKGROUND: Exposure to endocrine disruptors is unavoidable. Many such compounds are suspected to impact neurologic development of children, but most studies conducted have considered effects of individual chemicals in isolation. Because exposures co-occur, it is important to consider their health impacts in a single regression framework. METHODS: We applied Bayesian statistical tools (including shared mean and mixture priors for 25 unique chemicals) to study independent associations of endocrine disruptor biomarkers with autism spectrum disorder (ASD) (n = 491) and intellectual disability (n = 155), compared with 373 general population controls, in the Early Markers for Autism study. We measured biomarkers in maternal serum collected and stored from midpregnancy and considered them individually or as a class (i.e., summed polychlorinated biphenyls). We adjusted all models for original matching factors (child sex and month and year of birth), maternal age, maternal race/ethnicity, parity, and maternal education at the time samples were collected. We estimated the change in the odds of ASD or intellectual disability per 1 SD increase in the z-score of measured biomarker concentration for each chemical. RESULTS: Odds of ASD and intellectual disability did not change with increasing concentration for any specific endocrine disruptor. The effect estimates for each chemical were centered on or near an odds ratio of 1.00 in both models where we applied a shared mean or a mixture prior. CONCLUSION: Our mixtures analyses do not suggest an independent relationship with ASD or intellectual disability with any of the 25 chemicals examined together in this mixtures analysis.

Lung Cancer Risk Associated with Regulated and Unregulated Chrysotile Asbestos Fibers.

BACKGROUND: Regulation of asbestos fibers in the workplace is partly determined by which fibers can be visually counted. However, a majority of fibers are too short and thin to count this way and are, consequently, not subject to regulation. METHODS: We estimate lung cancer risk associated with asbestos fibers of varying length and width. We apply an order-constrained prior both to leverage external information from toxicological studies of asbestos health effects. This prior assumes that risk from asbestos fibers increases with increasing length and decreases with increasing width. RESULTS: When we apply a shared mean for the effect of all asbestos fiber exposure groups, the rate ratios for each fiber group per unit exposure appear mostly equal. Rate ratio estimates for fibers of diameter <0.25 µm and length <1.5 and 1.5-5.0 µm are the most precise. When applying an order-constrained prior, we find that estimates of lung cancer rate ratio per unit of exposure to unregulated fibers 20-40 and >40 µm in the thinnest fiber group are similar in magnitude to estimates of risk associated with long fibers in the regulated fraction of airborne asbestos fibers. Rate ratio estimates for longer fibers are larger than those for shorter fibers, but thicker and thinner fibers do not differ as the toxicologically derived prior had expected. CONCLUSION: Credible intervals for fiber size-specific risk estimates overlap; thus, we cannot conclude that there are substantial differences in effect by fiber size. Nonetheless, our results suggest that some unregulated asbestos fibers may be associated with increased incidence of lung cancer.

A Time Series Analysis of Associations between Daily Temperature and Crime Events in Philadelphia, Pennsylvania.

Urban crime may be an important but overlooked public health impact of rising ambient temperatures. We conducted a time series analysis of associations between temperature and crimes in Philadelphia, PA, for years 2006-2015. We obtained daily crime data from the Philadelphia Police Department, and hourly temperature and dew point data from the National Centers for Environmental Information. We calculated the mean daily heat index and daily deviations from each year's seasonal mean heat index value. We used generalized additive models with a quasi-Poisson distribution, adjusted for day of the week, public holiday, and long-term trends and seasonality, to estimate relative rates (RR) and 95% confidence intervals. We found that the strongest associations were with violent crime and disorderly conduct. For example, relative to the median of the distribution of mean daily heat index values, the rate of violent crimes was 9% (95% CI 6-12%) higher when the mean daily heat index was at the 99th percentile of the distribution. There was a positive, linear relationship between deviations of the daily mean heat index from the seasonal mean and rates of violent crime and disorderly conduct, especially in cold months. Overall, these analyses suggest that disorderly conduct and violent crimes are highest when temperatures are comfortable, especially during cold months. This work provides important information regarding the temporal patterns of crime activity.

Hierarchical regression for analyses of multiple outcomes.

In cohort mortality studies, there often is interest in associations between an exposure of primary interest and mortality due to a range of different causes. A standard approach to such analyses involves fitting a separate regression model for each type of outcome. However, the statistical precision of some estimated associations may be poor because of sparse data. In this paper, we describe a hierarchical regression model for estimation of parameters describing outcome-specific relative rate functions and associated credible intervals. The proposed model uses background stratification to provide flexible control for the outcome-specific associations of potential confounders, and it employs a hierarchical "shrinkage" approach to stabilize estimates of an exposure's associations with mortality due to different causes of death. The approach is illustrated in analyses of cancer mortality in 2 cohorts: a cohort of dioxin-exposed US chemical workers and a cohort of radiation-exposed Japanese atomic bomb survivors. Compared with standard regression estimates of associations, hierarchical regression yielded estimates with improved precision that tended to have less extreme values. The hierarchical regression approach also allowed the fitting of models with effect-measure modification. The proposed hierarchical approach can yield estimates of association that are more precise than conventional estimates when one wishes to estimate associations with multiple outcomes.

The researcher and the consultant: from testing to probability statements.

In the first instalment of this series, Stang and Poole provided an overview of Fisher significance testing (ST), Neyman-Pearson null hypothesis testing (NHT), and their unfortunate and unintended offspring, null hypothesis significance testing. In addition to elucidating the distinction between the first two and the evolution of the third, the authors alluded to alternative models of statistical inference; namely, Bayesian statistics. Bayesian inference has experienced a revival in recent decades, with many researchers advocating for its use as both a complement and an alternative to NHT and ST. This article will continue in the direction of the first instalment, providing practicing researchers with an introduction to Bayesian inference. Our work will draw on the examples and discussion of the previous dialogue.

Examining the association of lung cancer and highly correlated fibre size-specific asbestos exposures with a hierarchical Bayesian model.

BACKGROUND: Asbestos is a known carcinogen. However, little is known about the differential effects of size-specific asbestos fibres. Previous research has examined the relationship with lung cancer of each fibre group in the absence of others. Attempts to model all fibre groups within a single regression model have failed due to high correlations across fibre size groups. METHODS: We compare results from frequentist models for individual fibre size groups, and a hierarchical Bayesian model that included all fibre groups to estimate the relationship of size-specific asbestos fibre groups to lung cancer mortality. The hierarchical model assumes partial exchangeability of the effects of size-specific asbestos fibre groups to lung cancer, and is capable of handling the strong correlation of the exposure data. RESULTS: When fibre groups are modelled independently with a frequentist model, there appears to be an increase in the dose-response with increasing fibre size. However, when subject to a hierarchical structure, this trend vanishes, and the effects of distinct fibre groups appear largely similar. CONCLUSIONS: This is the first occasion where distinct asbestos fibre groups have been assessed in a single regression model; however, even the use of a hierarchical modelling structure does not appear to overcome all the statistical fluctuations arising from the high correlations across fibre groups. We believe these results should be compared with other occupational cohorts with similar fibre group information. Finally, results for the smallest fibre group may be suggestive of a carcinogenic potential for nanofibres.

Prenatal Metal Exposures and Child Social Responsiveness Scale Scores in 2 Prospective Studies.

Background: Prenatal exposure to metals is hypothesized to be associated with child autism. We aim to investigate the joint and individual effects of prenatal exposure to urine metals including lead (Pb), mercury (Hg), manganese (Mn), and selenium (Se) on child Social Responsiveness Scale (SRS) scores. Methods: We used data from 2 cohorts enriched for likelihood of autism spectrum disorder (ASD): Early Autism Risk Longitudinal Investigation (EARLI) and the Markers of Autism Risk in Babies-Learning Early Signs (MARBLES) studies. Metal concentrations were measured in urine collected during pregnancy. We used Bayesian Kernel Machine Regression and linear regression models to investigate both joint and independent associations of metals with SRS Z-scores in each cohort. We adjusted for maternal age at delivery, interpregnancy interval, maternal education, child race/ethnicity, child sex, and/or study site. Results: The final analytic sample consisted of 251 mother-child pairs. When Pb, Hg, Se, and Mn were at their 75th percentiles, there was a 0.03 increase (95% credible interval [CI]: -0.11, 0.17) in EARLI and 0.07 decrease (95% CI: -0.29, 0.15) in MARBLES in childhood SRS Z-scores, compared to when all 4 metals were at their 50th percentiles. In both cohorts, increasing concentrations of Pb were associated with increasing values of SRS Z-scores, fixing the other metals to their 50th percentiles. However, all the 95% credible intervals contained the null. Conclusions: There were no clear monotonic associations between the overall prenatal metal mixture in pregnancy and childhood SRS Z-scores at 36 months. There were also no clear associations between individual metals within this mixture and childhood SRS Z-scores at 36 months. The overall effects of the metal mixture and the individual effects of each metal within this mixture on offspring SRS Z-scores might be heterogeneous across child sex and cohort. Further studies with larger sample sizes are warranted.

Prenatal Phenol and Paraben Exposures and Adverse Birth Outcomes: A Prospective Analysis of U.S. Births.

BACKGROUND: Synthetic chemicals are increasingly being recognized for potential independent contributions to preterm birth (PTB) and low birth weight (LBW). Bisphenols, parabens, and triclosan are consumer product chemicals that act via similar mechanisms including estrogen, androgen, and thyroid disruption and oxidative stress. Multiple cohort studies have endeavored to examine effects on birth outcomes, and systematic reviews have been limited due to measurement of 1-2 spot samples during pregnancy and limited diversity of populations. OBJECTIVE: To study the effects of prenatal phenols and parabens on birth size and gestational age (GA) in 3,619 mother-infant pairs from 11 cohorts in the NIH Environmental influences on Child Health Outcomes program. RESULTS: While many associations were modest and statistically imprecise, a 1-unit increase in log10 pregnancy averaged concentration of benzophenone-3 and methylparaben were associated with decreases in birthweight, birthweight adjusted for gestational age and SGA. Increases in the odds of being SGA were 29% (95% CI: 5%, 58%) and 32% (95% CI: 3%, 70%), respectively. Bisphenol S in third trimester was also associated with SGA (OR 1.52, 95% CI 1.08, 2.13). Associations of benzophenone-3 and methylparaben with PTB and LBW were null. In addition, a 1-unit increase in log10 pregnancy averaged concentration of 2,4-dichlorophenol was associated with 43% lower (95% CI: -67%, -2%) odds of low birthweight; the direction of effect was the same for the highly correlated 2,5-dichlorophenol, but with a smaller magnitude (-29%, 95% CI: -53%, 8%). DISCUSSION: In a large and diverse sample generally representative of the United States, benzophenone-3 and methylparaben were associated with lower birthweight as well as birthweight adjusted for gestational age and higher odds of SGA, while 2,4-dichlorophenol. These associations with smaller size at birth are concerning in light of the known consequences of intrauterine growth restriction for multiple important health outcomes emerging later in life.

Prenatal phthalate exposure and adverse birth outcomes in the USA: a prospective analysis of births and estimates of attributable burden and costs.

BACKGROUND: Phthalates are synthetic chemicals widely used in consumer products and have been identified to contribute to preterm birth. Existing studies have methodological limitations and potential effects of di-2-ethylhexyl phthalate (DEHP) replacements are poorly characterised. Attributable fractions and costs have not been quantified, limiting the ability to weigh trade-offs involved in ongoing use. We aimed to leverage a large, diverse US cohort to study associations of phthalate metabolites with birthweight and gestational age, and estimate attributable adverse birth outcomes and associated costs. METHODS: In this prospective analysis we used extant data in the US National Institutes of Health Environmental influences on Child Health Outcomes (ECHO) Program from 1998 to 2022 to study associations of 20 phthalate metabolites with gestational age at birth, birthweight, birth length, and birthweight for gestational age z-scores. We also estimated attributable adverse birth outcomes and associated costs. Mother-child dyads were included in the study if there were one or more urinary phthalate measurements during the index pregnancy; data on child's gestational age and birthweight; and singleton delivery. FINDINGS: We identified 5006 mother-child dyads from 13 cohorts in the ECHO Program. Phthalic acid, diisodecyl phthalate (DiDP), di-n-octyl phthalate (DnOP), and diisononyl phthalate (DiNP) were most strongly associated with gestational age, birth length, and birthweight, especially compared with DEHP or other metabolite groupings. Although DEHP was associated with preterm birth (odds ratio 1·45 [95% CI 1·05-2·01]), the risks per log10 increase were higher for phthalic acid (2·71 [1·91-3·83]), DiNP (2·25 [1·67-3·00]), DiDP (1·69 [1·25-2·28]), and DnOP (2·90 [1·96-4·23]). We estimated 56 595 (sensitivity analyses 24 003-120 116) phthalate-attributable preterm birth cases in 2018 with associated costs of US$3·84 billion (sensitivity analysis 1·63- 8·14 billion). INTERPRETATION: In a large, diverse sample of US births, exposure to DEHP, DiDP, DiNP, and DnOP were associated with decreased gestational age and increased risk of preterm birth, suggesting substantial opportunities for prevention. This finding suggests the adverse consequences of substitution of DEHP with chemically similar phthalates and need to regulate chemicals with similar properties as a class. FUNDING: National Institutes of Health.

Prenatal Exposure to Nonpersistent Environmental Chemicals and Postpartum Depression.

Importance: Postpartum depression (PPD) affects up to 20% of childbearing individuals, and a significant limitation in reducing its morbidity is the difficulty in modifying established risk factors. Exposure to synthetic environmental chemicals found in plastics and personal care products, such as phenols, phthalates, and parabens, are potentially modifiable and plausibly linked to PPD and have yet to be explored. Objective: To evaluate associations of prenatal exposure to phenols, phthalates, parabens, and triclocarban with PPD symptoms. Design, Setting, and Participants: This was a prospective cohort study from 5 US sites, conducted from 2006 to 2020, and included pooled data from 5 US birth cohorts from the National Institutes of Health Environmental Influences on Child Health Outcomes (ECHO) consortium. Participants were pregnant individuals with data on urinary chemical concentrations (phenols, phthalate metabolites, parabens, or triclocarban) from at least 1 time point in pregnancy and self-reported postnatal depression screening assessment collected between 2 weeks and 12 months after delivery. Data were analyzed from February to May 2022. Exposures: Phenols (bisphenols and triclosan), phthalate metabolites, parabens, and triclocarban measured in prenatal urine samples. Main Outcomes and Measures: Depression symptom scores were assessed using the Edinburgh Postnatal Depression Scale (EPDS) or the Center for Epidemiologic Studies Depression Scale (CES-D), harmonized to the Patient-Reported Measurement Information System (PROMIS) Depression scale. Measures of dichotomous PPD were created using both sensitive (EPDS scores ≥10 and CES-D scores ≥16) and specific (EPDS scores ≥13 and CES-D scores ≥20) definitions. Results: Among the 2174 pregnant individuals eligible for analysis, nearly all (>99%) had detectable levels of several phthalate metabolites and parabens. PPD was assessed a mean (SD) of 3 (2.5) months after delivery, with 349 individuals (16.1%) and 170 individuals (7.8%) screening positive for PPD using the sensitive and specific definitions, respectively. Linear regression results of continuous PROMIS depression T scores showed no statistically significant associations with any chemical exposures. Models examining LMW and HMW phthalates and di (2-ethylhexyl) phthalate had estimates in the positive direction whereas all others were negative. A 1-unit increase in log-transformed LMW phthalates was associated with a 0.26-unit increase in the PROMIS depression T score (95% CI, -0.01 to 0.53; P = .06). This corresponded to an odds ratio (OR) of 1.08 (95% CI, 0.98-1.19) when modeling PPD as a dichotomous outcome and using the sensitive PPD definition. HMW phthalates were associated with increased odds of PPD (OR, 1.11; 95% CI, 1.00-1.23 and OR, 1.10; 95% CI, 0.96-1.27) for the sensitive and specific PPD definitions, respectively. Sensitivity analyses produced stronger results. Conclusions and Relevance: Phthalates, ubiquitous chemicals in the environment, may be associated with PPD and could serve as important modifiable targets for preventive interventions. Future studies are needed to confirm these observations.

Sensitivity analyses for sparse-data problems-using weakly informative bayesian priors.

Sparse-data problems are common, and approaches are needed to evaluate the sensitivity of parameter estimates based on sparse data. We propose a Bayesian approach that uses weakly informative priors to quantify sensitivity of parameters to sparse data. The weakly informative prior is based on accumulated evidence regarding the expected magnitude of relationships using relative measures of disease association. We illustrate the use of weakly informative priors with an example of the association of lifetime alcohol consumption and head and neck cancer. When data are sparse and the observed information is weak, a weakly informative prior will shrink parameter estimates toward the prior mean. Additionally, the example shows that when data are not sparse and the observed information is not weak, a weakly informative prior is not influential. Advancements in implementation of Markov Chain Monte Carlo simulation make this sensitivity analysis easily accessible to the practicing epidemiologist.