Visualizing External Validity: Graphical Displays to Inform the Extension of Treatment Effects from Trials to Clinical Practice

BACKGROUND

In the presence of effect measure modification, estimates of treatment effects from randomized controlled trials may not be valid in clinical practice settings. The development and application of quantitative approaches for extending treatment effects from trials to clinical practice settings is an active area of research.

METHODS

In this article, we provide researchers with a practical roadmap and four visualizations to assist in variable selection for models to extend treatment effects observed in trials to clinical practice settings and to assess model specification and performance. We apply this roadmap and visualizations to an example extending the effects of adjuvant chemotherapy (5-fluorouracil vs. plus oxaliplatin) for colon cancer from a trial population to a population of individuals treated in community oncology practices in the United States.

RESULTS

The first visualization screens for potential effect measure modifiers to include in models extending trial treatment effects to clinical practice populations. The second visualization displays a measure of covariate overlap between the clinical practice populations and the trial population. The third and fourth visualizations highlight considerations for model specification and influential observations. The conceptual roadmap describes how the output from the visualizations helps interrogate the assumptions required to extend treatment effects from trials to target populations.

CONCLUSIONS

The roadmap and visualizations can inform practical decisions required for quantitatively extending treatment effects from trials to clinical practice settings.

Individual and joint effects of prenatal PM2.5 and maternal stress on child temperament

Prenatal fine particulate matter (PM2.5) and maternal psychological functioning have been associated with child cognitive outcomes, though their independent and joint impacts on earlier behavioral outcomes remains less studied. We used data from 382 mother-child pairs from a prospective birth cohort in Mexico City. Temperament was measured at 24 months using the Carey Toddler Temperament Scale (TTS). Exploratory factor analysis (EFA) was used to update the factor structure of the TTS. During pregnancy, mothers completed the Crisis in Family Systems-Revised, Edinburgh Depression Scale, pregnancy-specific anxiety scale, and the Perceived Stress Scale. Pregnancy PM2.5 was assessed using estimates from a satellite-based exposure model. We assessed the association between prenatal maternal stress and PM2.5 on temperament, in both independent and joint models. Quantile g-computation was used to estimate the joint associations. Models were adjusted for maternal age, SES, education, child sex, and child age. In EFA, we identified three temperament factors related to effortful control, extraversion, and negative affect. Our main results showed that higher levels of PM2.5 and several of the maternal psychological functioning measures were related to both effortful control and negative affect in the child, both individually and as a mixture. For instance, a one quartile increase in the prenatal mixture was associated with higher negative affect scores in the child (0.34, 95% CI: 0.16, 0.53). We observed modification of these associations by maternal SES, with associations seen only among lower SES participants for both effortful control (-0.45, 95% CI: -0.70, -0.20) and negative affect outcomes (0.60, 95% CI: 0.35, 0.85). Prenatal PM2.5 and maternal psychological functioning measures were associated with toddler temperament outcomes, providing evidence for impacts of chemical and non-chemical stressors on early child health.

Exposure to indoor light at night in relation to multiple dimensions of sleep health: findings from the Sister Study

STUDY OBJECTIVE

To examine the association between light at night (LAN) and multiple sleep health dimensions.

METHODS

Among 47 765 Sister Study participants, indoor LAN (TV on in the room, light(s) on in room, light from outside the room, nightlight, no light) and sleep dimensions were self-reported at baseline (2003-2009). We used Poisson regression with robust variance to estimate adjusted prevalence ratios (PR) and 95% confidence intervals (CI) for the cross-sectional associations between LAN and short sleep duration (<7 hours/night), insomnia symptoms (difficulty falling or staying asleep), frequent napping (≥3 naps/week), inconsistent sleep/wake time (differed day-to-day and week-to-week), sleep debt (≥2 hours between longest and shortest duration), recent sleep medication use, and a cumulative poor sleep score (≥3 poor sleep dimensions). Population-attributable risks (PARs) were determined for any light exposure vs. none by race/ethnicity.

RESULTS

Compared to sleeping with no light in the bedroom, sleeping with a TV on was associated with a higher prevalence of most dimensions of poor sleep (e.g. short sleep duration: PR = 1.38, 95% CI: 1.32 to 1.45; inconsistent sleep/wake time: PR = 1.55, 95% CI: 1.44 to 1.66; sleep debt: PR = 1.36, 95% CI: 1.29 to 1.44; poor sleep score: PR = 1.58, 95% CI: 1.48-1.68). PARs tended to be higher for non-Hispanic black women compared to non-Hispanic white women.

CONCLUSIONS

Sleeping with a TV on was associated with poor sleep health among US women, and non-Hispanic black women may be disproportionately burdened.

Personal care product use patterns in association with phthalate and replacement biomarkers across pregnancy

BACKGROUND

Humans are exposed to phthalates, a class of non-persistent chemicals, through multiple products, including personal care and cosmetics. Associations between specific phthalates and product use have been inconsistent. However, determining these connections could provide avenues for exposure reduction.

OBJECTIVE

Examine the association between patterns of personal care product use and associations with phthalate and replacement biomarkers.

METHODS

In the Human Placenta and Phthalates Study, 303 women were enrolled in early pregnancy and followed for up to 8 visits across gestation. At each visit, women completed a questionnaire about product use in the prior 24 hours and contributed urine samples, subsequently analyzed for 18 phthalate and replacement metabolites. At early, mid-, and late pregnancy, questionnaire responses were condensed and repeated metabolite concentrations were averaged. Latent class analysis (LCA) was used to determine groups of women with similar use patterns, and weighted associations between group membership and biomarker concentrations were assessed.

RESULTS

LCA sorted women into groups which largely corresponded to: (1) low fragranced product use (16-23% of women); (2) fragranced product and low body wash use (22-26%); 3) fragranced product and low bar soap use (26-51%); and (4) low product use (7-34%). Monoethyl phthalate (MEP) urinary concentrations were 7-10% lower and concentrations of summed di(2-ethylhexyl) terephthalate metabolites were 15-21% lower among women in the "low fragranced product use" group compared to the population mean. Few other consistent associations between group and biomarker concentrations were noted.

IMPACT STATEMENT

Personal care products and cosmetics are a known exposure source for phthalates and potentially represent one of the most accessible intervention targets for exposure reduction. However, in this analysis accounting for concurrent use and fragranced status of products, we did not find any use patterns that corresponded to universally lower levels.

Use of personal care product mixtures and incident hormone-sensitive cancers in the Sister Study: A U.S.-wide prospective cohort

BACKGROUND

Personal care products (PCPs), a source of endocrine-disrupting chemical exposure, may be associated with the risk of hormone-sensitive cancers. Few studies have investigated associations for PCP use with the incidence of hormone-sensitive cancers or considered the joint effect of multiple correlated PCPs. We examined associations between frequently used, or "everyday", PCPs and incident cancers of the breast, ovary, and uterus with a fucus on the joint effect of multiple product exposure.

METHODS

Sister Study participants (n=49 899) self-reported frequency of use in the year before enrollment (2003-2009) for 41 PCPs. Using five-level frequency categories based on questionnaire options, hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated for the associations between multiple PCP use and incident breast, ovarian, and uterine cancer using quantile-based g-computation with Cox proportional hazards regression as the underlying model. Multiple PCP use was examined using groupings (beauty, hygiene, and skincare products) determined by both a priori knowledge and Spearman correlation coefficients for co-occurring product use. Associations between individual PCPs and the three cancers were also examined using Cox proportional hazards models coupling with Benjamini-Hochberg procedure for multiple comparisons.

RESULTS

Over an average of 11.6 years, 4 226 breast, 277 ovarian, and 403 uterine cancer cases were identified. Positive associations were observed between the hygiene mixture and ovarian cancer (HR=1.35, 95%CI=1.00, 1.83) and the beauty mixture with postmenopausal breast cancer (HR=1.08, 95%CI=1.01, 1.16). Additionally, we observed an inverse association between the skincare mixture and breast cancer (HR=0.91, 95%CI=0.83, 0.99). No significant associations were observed for individual products after corrected for multiple comparison.

CONCLUSIONS

Findings from this multi-product, joint-effect approach contribute to the growing body of evidence for associations between PCPs and breast cancer and provides novel information on ovarian and uterine cancer.

Gender differences in the surgical management of trachomatous trichiasis: an exploratory analysis of global trachoma survey data, 2015-2019

BACKGROUND

Trachomatous trichiasis (TT) is a painful, potentially blinding eye condition that can be managed through epilation or surgery. Women are affected by TT approximately twice as often as men and are believed to face gendered barriers to receiving surgical care to prevent vision loss.

METHODS

We used data from 817 cross-sectional surveys conducted during 2015-2019 in 20 African countries to estimate the prevalence difference (PD) between female and male eyes for four outcomes potentially indicating gender-related differences in TT management: (1) received surgery and developed postoperative TT (PTT), (2) never offered surgery, (3) offered surgery but declined it, and (4) offered epilation but never offered surgery.

RESULTS

The prevalence was modestly elevated among female eyes compared with male eyes for having PTT (PD:1.8 [95% confidence limits (CL): 0.6, 3.0]) and having declined surgery for the eye (PD: 6.2 [95% CL: 1.8, 10.7]). The proportion offered epilation was similar by gender (PD:0.5 [95% CL: -0.4, 1.3]), while never having been offered surgery was somewhat more prevalent among male eyes (PD: -2.1 [95% CL: -3.5, -0.7]).

CONCLUSIONS

Our results suggest potential gender differences in TT management. More research is needed to determine the causes and implications of the observed differences.

Racial and Ethnic Disparities in Phthalate Exposure and Preterm Birth: A Pooled Study of Sixteen U.S. Cohorts

BACKGROUND

Phthalate exposures are ubiquitous during pregnancy and may contribute to racial and ethnic disparities in preterm birth.

OBJECTIVES

We investigated race and ethnicity in the relationship between biomarkers of phthalate exposure and preterm birth by examining: a) how hypothetical reductions in racial and ethnic disparities in phthalate metabolites might reduce the probability of preterm birth; and b) exposure-response models stratified by race and ethnicity.

METHODS

We pooled individual-level data on 6,045 pregnancies from 16 U.S. cohorts. We investigated covariate-adjusted differences in nine urinary phthalate metabolite concentrations by race and ethnicity [non-Hispanic White (White, 43%), non-Hispanic Black (Black, 13%), Hispanic/Latina (38%), and Asian/Pacific Islander (3%)]. Using g-computation, we estimated changes in the probability of preterm birth under hypothetical interventions to eliminate disparities in levels of urinary phthalate metabolites by proportionally lowering average concentrations in Black and Hispanic/Latina participants to be approximately equal to the averages in White participants. We also used race and ethnicity-stratified logistic regression to characterize associations between phthalate metabolites and preterm birth.

RESULTS

In comparison with concentrations among White participants, adjusted mean phthalate metabolite concentrations were consistently higher among Black and Hispanic/Latina participants by 23%-148% and 4%-94%, respectively. Asian/Pacific Islander participants had metabolite levels that were similar to those of White participants. Hypothetical interventions to reduce disparities in metabolite mixtures were associated with lower probabilities of preterm birth for Black [13% relative reduction; 95% confidence interval (CI): - 34 % , 8.6%] and Hispanic/Latina (9% relative reduction; 95% CI: - 19 % , 0.8%) participants. Odds ratios for preterm birth in association with phthalate metabolites demonstrated heterogeneity by race and ethnicity for two individual metabolites (mono-n-butyl and monoisobutyl phthalate), with positive associations that were larger in magnitude observed among Black or Hispanic/Latina participants.

CONCLUSIONS

Phthalate metabolite concentrations differed substantially by race and ethnicity. Our results show hypothetical interventions to reduce population-level racial and ethnic disparities in biomarkers of phthalate exposure could potentially reduce the probability of preterm birth. https://doi.org/10.1289/EHP12831.

Identifying and estimating effects of sustained interventions under parallel trends assumptions

Many research questions in public health and medicine concern sustained interventions in populations defined by substantive priorities. Existing methods to answer such questions typically require a measured covariate set sufficient to control confounding, which can be questionable in observational studies. Differences-in-differences rely instead on the parallel trends assumption, allowing for some types of time-invariant unmeasured confounding. However, most existing difference-in-differences implementations are limited to point treatments in restricted subpopulations. We derive identification results for population effects of sustained treatments under parallel trends assumptions. In particular, in settings where all individuals begin follow-up with exposure status consistent with the treatment plan of interest but may deviate at later times, a version of Robins' g-formula identifies the intervention-specific mean under stable unit treatment value assumption, positivity, and parallel trends. We develop consistent asymptotically normal estimators based on inverse-probability weighting, outcome regression, and a double robust estimator based on targeted maximum likelihood. Simulation studies confirm theoretical results and support the use of the proposed estimators at realistic sample sizes. As an example, the methods are used to estimate the effect of a hypothetical federal stay-at-home order on all-cause mortality during the COVID-19 pandemic in spring 2020 in the United States.

Iterative Causal Forest: A Novel Algorithm for Subgroup Identification

Precisely and efficiently identifying subgroups with heterogeneous treatment effects (HTEs) in real-world evidence studies remains a challenge. Based on the causal forest (CF) method, we developed an iterative CF (iCF) algorithm to identify HTEs in subgroups defined by important variables. Our method iteratively grows different depths of the CF with important effect modifiers, performs plurality votes to obtain decision trees (subgroup decisions) for a family of CFs with different depths, then finds the cross-validated subgroup decision that best predicts the treatment effect as a final subgroup decision. We simulated 12 different scenarios and showed that the iCF outperformed other machine learning methods for interaction/subgroup identification in the majority of scenarios assessed. Using a 20% random sample of fee-for-service Medicare beneficiaries initiating sodium-glucose cotransporter-2 inhibitors (SGLT2i) or glucagon-like peptide-1 receptor agonists (GLP1RA), we implemented the iCF to identify subgroups with HTEs for hospitalized heart failure. Consistent with previous studies suggesting patients with heart failure benefit more from SGLT2i, iCF successfully identified such a subpopulation with HTEs and additive interactions. The iCF is a promising method for identifying subgroups with HTEs in real-world data where the potential for unmeasured confounding can be limited by study design.

Toxic metal mixtures in private well water and increased risk for preterm birth in North Carolina

BACKGROUND

Prenatal exposure to metals in private well water may increase the risk of preterm birth (PTB) (delivery < 37 weeks' gestation). In this study, we estimated associations between arsenic, manganese, lead, cadmium, chromium, copper, and zinc concentrations in private well water and PTB incidence in North Carolina (NC).

METHODS

Birth certificates from 2003-2015 (n = 1,329,071) were obtained and pregnancies were assigned exposure using the mean concentration and the percentage of tests above the maximum contaminant level (MCL) for the census tract of each individuals' residence at the time of delivery using the NCWELL database (117,960 well water tests from 1998-2019). We evaluated associations between single metals and PTB using adjusted logistic regression models. Metals mixtures were assessed using quantile-based g-computation.

RESULTS

Compared with those in other census tracts, individuals residing in tracts where > 25% of tests exceeded the MCL for lead (aOR 1.10, 95%CI 1.02,1.18) or cadmium (aOR 1.11, 95% CI 1.00,1.23) had an increased odds of PTB. Conversely, those residing in areas with > 25% MCL for zinc (aOR 0.77 (95% CI: 0.56,1.02) and copper (aOR 0.53 (95% CI: 0.13,1.34)) had a reduced odds of PTB. A quartile increase in the concentrations of a mixture of lead, cadmium, and chromium was associated with a small increased odds for PTB (aOR 1.02, 95% CI 1.01, 1.03). This metal mixture effect was most pronounced among American Indian individuals (aOR per quartile increase in all metals: 1.19 (95% CI 1.06,1.34)).

CONCLUSIONS

In a large study population of over one million births, lead and cadmium were found to increase the risk of PTB individually and in a mixture, with additional mixtures-related impacts estimated from co-exposure with chromium. This study highlights critical racial and ethnic health disparities in relation to private well water thereby emphasizing the urgent need for improved private well water quality to protect vulnerable populations.

Variability and Longitudinal Trajectories of Phthalate and Replacement Biomarkers across Pregnancy in the Human Placenta and Phthalates Study

Human exposure to phthalates is widespread, but assessment of variability across pregnancy has been hampered by short half-lives of phthalate biomarkers and a few repeated measures in prior studies. We aimed to characterize the variability and longitudinal profiles of phthalate and replacement biomarkers across pregnancy. Within the Human Placenta and Phthalates Study, 303 pregnant women provided urine samples at up to 8 visits across gestation. Concentrations of 14 metabolites of phthalates and 4 metabolites of replacements were quantified in each sample, and subject-specific averages within each trimester were calculated. We examined variability in individual biomarker concentrations across the 8 visits, within trimesters, and across trimester-specific averages using intraclass correlation coefficients (ICCs). To explore longitudinal exposure biomarker profiles, we applied group-based trajectory modeling to trimester-specific averages over pregnancy. Pooling multiple visits into trimester-specific averages improved the ICCs for all biomarkers. Most biomarkers generally showed stable concentrations across gestation, i.e., high-, medium-, and low-concentration profiles, with small proportions of participants falling into the "high"-exposure groups. Variability over pregnancy is likely attributable to random fluctuations around a baseline exposure rather than true changes in concentrations over time.

Differential Employment Quality and Educational Inequities in Mental Health: A Causal Mediation Analysis

BACKGROUND

In the United States, inequities in mental distress between those more and less educated have widened over recent years. Employment quality, a multidimensional construct reflecting the relational and contractual features of employer-employee relationships, may mediate this inequity throughout adulthood, yet no study has examined the extent of this mediation in the United States, or how it varies across racialized and gendered populations.

METHODS

Using the information on working-age adults from the 2001 to 2019 Panel Study of Income Dynamics, we construct a composite measure of employment quality via principal component analysis. Using this measure and the parametric mediational g-formula, we then estimate randomized interventional analogs for natural direct and indirect effects of low baseline educational attainment (≤high school: no/yes) on the end-of-follow-up prevalence of moderate mental distress (Kessler-6 Score ≥5: no/yes) overall and within subgroups by race and gender.

RESULTS

We estimate that low educational attainment would result in a 5.3% greater absolute prevalence of moderate mental distress at the end of follow-up (randomized total effect: 5.3%, 95% CI = 2.2%, 8.4%), with approximately 32% of this effect mediated by differences in employment quality (indirect effect: 1.7%, 95% CI = 1.0%, 2.5%). The results of subgroup analyses across race and gender are consistent with the hypothesis of mediation by employment quality, though not when selecting on full employment (indirect effect: 0.6%, 95% CI = -1.0%, 2.6%).

CONCLUSIONS

We estimate that approximately one-third of US educational inequities in mental distress may be mediated by differences in employment quality.

Midpregnancy Phthalate and Phenol Biomarkers in Relation to Infant Body Composition: The Healthy Start Prospective Cohort

BACKGROUND

Gestational phthalate and phenol exposure disrupts adipogenesis, contributing to obesity in mice. Whether gestational phthalate or phenol exposure is associated with infant body composition has not been investigated in humans.

OBJECTIVE

We examined associations between biomarkers of phthalate and phenol exposure in midpregnancy and infant size and body composition at birth and at 5 months of age.

METHODS

Analyses were conducted among 438 infants from the Healthy Start prospective pregnancy cohort. Sixteen phthalate and phenol biomarkers were quantified in spot urine samples collected at 24-28 wk of gestation. Infant outcomes measured at birth and at 5 months of age included size [weight (in grams)] and body composition [fat and lean masses (in grams); percentage fat mass]. Single- (linear) and multipollutant (quantile g-computation) models were used to estimate associations of phthalate and phenol biomarkers with infant outcomes at birth and at 5 months of age. Models were adjusted for sociodemographics, sample collection timing, and lifestyle factors and used to examine for effect modification by infant sex.

RESULTS

In single-pollutant models, mono-benzyl phthalate and di-n -butyl phthalate were inversely associated with percentage fat mass [β : - 0.49 (95% CI: - 0.91 , - 0.08 ) and - 0.51 (95% CI: - 1.02 , 0.01), respectively] in male but not female infants at birth. Similar, but less precise, associations were observed at 5 months of age. In multipollutant models, a 1-quartile increase in the phthalate and phenol biomarker mixture was inversely associated with percentage fat mass at birth [- 1.06 (95% CI: - 2.21 , 0.1)] and at 5 months of age [- 2.14 (95% CI: - 3.88 , - 0.39 )] among males, but associations were null among females [0.48 (95% CI: - 0.78 , 1.75) and - 0.64 (95% CI: - 2.68 , 1.41), respectively]. Similar associations were observed with infant weight.

CONCLUSION

In this U.S.-based prospective cohort, gestational phthalate and phenol biomarkers were inversely associated with infant weight and fat mass, particularly in males. https://doi.org/10.1289/EHP12500.

Maternal serum per- and polyfluoroalkyl substances during pregnancy and breastfeeding duration

Per- and polyfluoroalkyl substances (PFAS) are endocrine-disrupting chemicals that may affect breastfeeding duration. We examined associations between maternal PFAS concentrations during pregnancy and breastfeeding cessation. We investigated potential effect modification by parity status.

Methods

Among 555 women enrolled in the Healthy Start study (2009-2014), we quantified maternal serum concentrations of 5 PFAS during mid- to late-pregnancy (mean 27 weeks of gestation). Participants self-reported their breastfeeding practices through 18-24 months postnatally. Among all participants and stratified by parity, we estimated associations between maternal PFAS concentrations and breastfeeding discontinuation by 3 and 6 months, using Poisson regression, and breastfeeding duration, using Cox regression.

Results

Median PFAS concentrations were similar to those in the general US population. Associations between PFAS and breastfeeding duration differed by parity status. After adjusting for covariates, among primiparous women, associations between PFAS and breastfeeding cessation by 3 and 6 months were generally null, with some inverse associations. Among multiparous women, there were positive associations between perfluorohexane sulfonate, perfluorooctane sulfonate, perfluorooctanoate (PFOA), and perfluorononanoate and breastfeeding cessation by 3 and 6 months. For example, per ln-ng/mL increase in PFOA, the risk ratio for breastfeeding discontinuation by 6 months was 1.45 (95% confidence interval, 1.18, 1.78). Hazard ratios reflected similar patterns between PFAS and breastfeeding duration.

Conclusions

Among primiparous women, we did not find evidence for associations between PFAS concentrations and breastfeeding duration. In contrast, among multiparous women, PFAS serum concentrations were generally inversely associated with breastfeeding duration, though estimates may be biased due to confounding by unmeasured previous breastfeeding.

Prenatal Exposure to Toxic Metals and Neural Tube Defects: A Systematic Review of the Epidemiologic Evidence

BACKGROUND

Neural tube defects (NTDs) affect > 300,000 pregnancies worldwide annually. Few nongenetic factors, other than folate deficiency, have been identified that may provide intervenable solutions to reduce the burden of NTDs. Prenatal exposure to toxic metals [arsenic (As), cadmium (Cd), mercury (Hg), manganese (Mn) and lead (Pb)] may increase the risk of NTDs. Although a growing epidemiologic literature has examined associations, to our knowledge no systematic review has been conducted to date.

OBJECTIVE

Through adaptation of the Navigation Guide systematic review methodology, we aimed to answer the question "does exposure to As, Cd, Hg, Mn, or Pb during gestation increase the risk of NTDs?" and to assess challenges to evaluating this question given the current evidence.

METHODS

We selected available evidence on prenatal As, Cd, Hg, Mn, or Pb exposure and risk of specific NTDs (e.g., spina bifida, anencephaly) or all NTDs via a comprehensive search across MEDLINE, Embase, Web of Science, and TOXLINE databases and applied inclusion/exclusion criteria. We rated the quality and strength of the evidence for each metal. We applied a customized risk of bias protocol and evaluated the sufficiency of evidence of an effect of each metal on NTDs.

RESULTS

We identified 30 studies that met our criteria. Risk of bias for confounding and selection was high in most studies, but low for missing data. We determined that, although the evidence was limited, the literature supported an association between prenatal exposure to Hg or Mn and increased risk of NTDs. For the remaining metals, the evidence was inadequate to establish or rule out an effect.

CONCLUSION

The role of gestational As, Cd, or Pb exposure in the etiology of NTDs remains unclear and warrants further investigation in high-quality studies, with a particular focus on controlling confounding, mitigating selection bias, and improving exposure assessment. https://doi.org/10.1289/EHP11872.

Prenatal organophosphate ester exposure and executive function in Norwegian preschoolers

Organophosphate esters (OPEs) are ubiquitous chemicals, used as flame retardants and plasticizers. OPE usage has increased over time as a substitute for other controlled compounds. This study investigates the impact of prenatal OPE exposure on executive function (EF) in preschoolers.

Methods

We selected 340 preschoolers from the Norwegian Mother, Father, and Child Cohort Study. Diphenyl-phosphate (DPhP), di-n-butyl-phosphate (DnBP), bis(2-butoxyethyl) phosphate (BBOEP), and bis(1,3-dichloro-2-propyl) phosphate (BDCIPP) were measured in maternal urine. EF was measured using the Behavior Rating Inventory of Executive Functioning-Preschool (BRIEF-P) and the Stanford-Binet fifth edition (SB-5). EF scores were scaled so a higher score indicated worse performance. We estimated exposure-outcome associations and evaluated modification by child sex using linear regression.

Results

Higher DnBP was associated with lower EF scores across multiple rater-based domains. Higher DPhP and BDCIPP were associated with lower SB-5 verbal working memory (β = 0.49, 95% CI = 0.12, 0.87; β = 0.53, 95% CI = 0.08, 1.02), and higher BBOEP was associated with lower teacher-rated inhibition (β = 0.34, 95% CI = 0.01, 0.63). DPhP was associated with lower parent-reported BRIEF-P measures in boys but not girls [inhibition: boys: 0.37 (95% CI = 0.03, 0.93); girls: -0.48 (95% CI = -1.27, 0.19); emotional control: boys: 0.44 (95% CI = -0.13, 1.26); girls: -0.83 (95% CI = -1.73, -0.00); working memory: boys: 0.49 (95% CI = 0.03, 1.08); girls: -0.40 (95% CI = -1.11, 0.36)]. Fewer sex interactions were observed for DnBP, BBOEP, and BDCIPP, with irregular patterns observed across EF domains.

Conclusions

We found some evidence prenatal OPE exposure may impact EF in preschoolers and variation in associations by sex.

Volatile Hydrocarbon Exposures and Incident Coronary Heart Disease Events: Up to Ten Years of Follow-up among Deepwater Horizon Oil Spill Workers

BACKGROUND

During the 2010 Deepwater Horizon (DWH) disaster, response and cleanup workers were potentially exposed to toxic volatile components of crude oil. However, to our knowledge, no study has examined exposure to individual oil spill-related chemicals in relation to cardiovascular outcomes among oil spill workers.

OBJECTIVES

Our aim was to investigate the association of several spill-related chemicals [benzene, toluene, ethylbenzene, xylene, n-hexane (BTEX-H)] and total hydrocarbons (THC) with incident coronary heart disease (CHD) events among workers enrolled in a prospective cohort.

METHODS

Cumulative exposures to THC and BTEX-H across the cleanup period were estimated via a job-exposure matrix that linked air measurement data with self-reported DWH spill work histories. We ascertained CHD events following each worker's last day of cleanup work as the first self-reported physician-diagnosed myocardial infarction (MI) or a fatal CHD event. We estimated hazard ratios (HR) and 95% confidence intervals for the associations of exposure quintiles (Q) with risk of CHD. We applied inverse probability weights to account for bias due to confounding and loss to follow-up. We used quantile g-computation to assess the joint effect of the BTEX-H mixture.

RESULTS

Among 22,655 workers with no previous MI diagnoses, 509 experienced an incident CHD event through December 2019. Workers in higher quintiles of each exposure agent had increased CHD risks in comparison with the referent group (Q1) of that agent, with the strongest associations observed in Q5 (range of HR = 1.14 - 1.44 ). However, most associations were nonsignificant, and there was no evidence of exposure-response trends. We observed stronger associations among ever smokers, workers with ≤ high school education, and workers with body mass index < 30   kg / m 2 . No apparent positive association was observed for the BTEX-H mixture.

CONCLUSIONS

Higher exposures to volatile components of crude oil were associated with modest increases in risk of CHD among oil spill workers, although we did not observe exposure-response trends. https://doi.org/10.1289/EHP11859.

Hurricane Florence and suicide mortality in North Carolina: a controlled interrupted time-series analysis

BACKGROUND

Natural disasters are associated with increased mental health disorders and suicidal ideation; however, associations with suicide deaths are not well understood. We explored how Hurricane Florence, which made landfall in September 2018, may have impacted suicide deaths in North Carolina (NC).

METHODS

We used publicly available NC death records data to estimate associations between Hurricane Florence and monthly suicide death rates using a controlled, interrupted time series analysis. Hurricane exposure was determined by using county-level support designations from the Federal Emergency Management Agency. We examined effect modification by sex, age group, and race/ethnicity.

RESULTS

8363 suicide deaths occurred between January 2014 and December 2019. The overall suicide death rate in NC between 2014 and 2019 was 15.53 per 100 000 person-years (95% CI 15.20 to 15.87). Post-Hurricane, there was a small, immediate increase in the suicide death rate among exposed counties (0.89/100 000 PY; 95% CI -2.69 to 4.48). Comparing exposed and unexposed counties, there was no sustained post-Hurricane Florence change in suicide death rate trends (0.02/100 000 PY per month; 95% CI -0.33 to 0.38). Relative to 2018, NC experienced a statewide decline in suicides in 2019. An immediate increase in suicide deaths in Hurricane-affected counties versus Hurricane-unaffected counties was observed among women, people under age 65 and non-Hispanic black individuals, but there was no sustained change in the months after Hurricane Florence.

CONCLUSIONS

Although results did not indicate a strong post-Hurricane Florence impact on suicide rates, subgroup analysis suggests differential impacts of Hurricane Florence on several groups, warranting future follow-up.

Estimating Causal Effects of Interventions on Early-life Environmental Exposures Using Observational Data

PURPOSE OF REVIEW

We discuss how epidemiologic studies have used observational data to estimate the effects of potential interventions on early-life environmental exposures. We summarize the value of posing questions about interventions, how a group of techniques known as "g-methods" can provide advantages for estimating intervention effects, and how investigators have grappled with the strong assumptions required for causal inference.

RECENT FINDINGS

We identified nine studies that estimated health effects of hypothetical interventions on early-life environmental exposures. Of these, six examined air pollution. Interventions evaluated by these studies included setting exposure levels at a specific value, shifting exposure distributions, and limiting exposure levels to less than a threshold value. Only one study linked exposure contrasts to a specific intervention on an exposure source, however. There is growing interest in estimating intervention effects of early-life environmental exposures, in part because intervention effects are directly related to possible public health actions. Future studies can build on existing work by linking research questions to specific hypothetical interventions that could reduce exposure levels. We discuss how framing questions around interventions can help overcome some of the barriers to causal inference and how advances related to machine learning may strengthen studies by sidestepping the overly restrictive assumptions of parametric regression models. By leveraging advancements in causal inference and exposure science, an intervention framework for environmental epidemiology can guide actionable solutions to improve children's environmental health.

Gestational organophosphate ester exposure and preschool attention-deficit/hyperactivity disorder in the Norwegian Mother, Father, and Child cohort study

BACKGROUND

Attention-deficit/hyperactivity-disorder (ADHD) is a leading neurodevelopmental disorder in children worldwide; however, few modifiable risk factors have been identified. Organophosphate esters (OPEs) are ubiquitous chemical compounds that are increasingly prevalent as a replacement for other regulated chemicals. Current research has linked OPEs to neurodevelopmental deficits. The purpose of this study was to assess gestational OPE exposure on clinically-assessed ADHD in children at age 3 years.

METHODS

In this nested case-control study within the Norwegian Mother, Father, and Child Cohort study, we evaluated the impact of OPE exposure at 17 weeks' gestation on preschool-age ADHD. Between 2007 and 2011, 260 ADHD cases were identified using the Preschool Age Psychiatric Assessment and compared to a birth-year-stratified control group of 549 children. We categorized bis(2-butoxyethyl) phosphate (BBOEP) and bis(1,3-dichloro-2-propyl) phosphate (BDCIPP) as values < limit of detection (LOD) (BBOEP N = 386, BDCIPP N = 632), ≥LOD but < limit of quantification (LOQ) (BBOEP N = 413; BDCIPP N = 75), or above LOQ (BBOEP N = 70; BDCIPP N = 102). Diphenyl phosphate (DPhP) and di-n-butyl phosphate (DnBP) were categorized as quartiles and also modeled with a log10 linear term. We estimated multivariable adjusted odds ratios (ORs) using logistic regression and examined modification by sex using an augmented product term approach.

RESULTS

Mothers in the 3rd DnBP quartile had 1.71 times the odds of having a child with ADHD compared to the 1st quartile (95%CI: 1.13, 2.58); a similar trend was observed for log10 DnBP and ADHD. Mothers with BDCIPP ≥ LOD but < LOQ had 1.39 times the odds of having a child with ADHD compared to those with BDCIPP < LOD (95%CI: 0.83, 2.31). Girls had lower odds of ADHD with increasing BBOEP exposure (log10 OR: 0.55 (95%CI: 0.37, 0.93), however boys had a weakly increased odds (log10 OR: 1.25 (95%CI: 0.74, 2.11) p-interaction = 0.01].

CONCLUSIONS

We found modest increased odds of preschool ADHD with higher DnBP and BDCIPP exposure.

How Effect Measure Choice Influences Minimally Sufficient Adjustment Sets for External Validity

Epidemiologic researchers generalizing or transporting effect estimates from a study to a target must account for effect measure modifiers (EMMs) on the scale of interest. Little attention is paid to how the EMMs required may vary depending on the mathematical nuances of each effect measure, however. We defined two types of EMM: marginal EMM, where the effect on the scale of interest differs across levels of a variable; and conditional EMM, where the effect differs conditional on other variables associated with the outcome. These types define three classes of variables: Class 1 (conditional EMM), Class 2 (marginal, but not conditional, EMM), or Class 3 (neither marginal nor conditional EMM). Class 1 variables are necessary to achieve a valid estimate of the RD in a target, while a RR requires Class 1 and Class 2 and an OR requires Class 1, Class 2, and Class 3 (i.e., all variables associated with the outcome). This does not mean that fewer variables are required for an externally valid RD (because variables may not modify effects on all scale) but does suggest researchers should consider the scale of the effect measure when identifying EMM necessary for an externally valid treatment effect estimate.

Exploring water, sanitation, and hygiene coverage targets for reaching and sustaining trachoma elimination: G-computation analysis

BACKGROUND

Trachoma is the leading infectious cause of blindness. To reduce transmission, water, sanitation, and hygiene (WaSH) improvements are promoted through a comprehensive public health strategy. Evidence supporting the role of WaSH in trachoma elimination is mixed and it remains unknown what WaSH coverages are needed to effectively reduce transmission.

METHODS/FINDINGS

We used g-computation to estimate the impact on the prevalence of trachomatous inflammation-follicular among children aged 1-9 years (TF1-9) when hypothetical WaSH interventions raised the minimum coverages from 5% to 100% for "nearby" face-washing water (<30 minutes roundtrip collection time) and adult latrine use in an evaluation unit (EU). For each scenario, we estimated the generalized prevalence difference as the TF1-9 prevalence under the intervention scenarios minus the observed prevalence. Data from 574 cross-sectional surveys conducted in 16 African and Eastern Mediterranean countries were included. Surveys were conducted from 2015-2019 with support from the Global Trachoma Mapping Project and Tropical Data. When modeling interventions among EUs that had not yet met the TF1-9 elimination target, increasing nearby face-washing water and latrine use coverages above 30% was generally associated with consistent decreases in TF1-9. For nearby face-washing water, we estimated a ≥25% decrease in TF1-9 at 65% coverage, with a plateau upon reaching 85% coverage. For latrine use, the estimated decrease in TF1-9 accelerated from 80% coverage upward, with a ≥25% decrease in TF1-9 by 85% coverage. Among EUs that had previously met the elimination target, results were inconclusive.

CONCLUSIONS

Our results support Sustainable Development Goal 6 and provide insight into potential WaSH-related coverage targets for trachoma elimination. Targets can be tested in future trials to improve evidence-based WaSH guidance for trachoma.

Fine particulate matter and incident coronary heart disease events up to 10 years of follow-up among Deepwater Horizon oil spill workers

BACKGROUND

During the 2010 Deepwater Horizon (DWH) disaster, in-situ burning and flaring were conducted to remove oil from the water. Workers near combustion sites were potentially exposed to burning-related fine particulate matter (PM2.5). Exposure to PM2.5 has been linked to increased risk of coronary heart disease (CHD), but no study has examined the relationship among oil spill workers.

OBJECTIVES

To investigate the association between estimated PM2.5 from burning/flaring of oil/gas and CHD risk among the DWH oil spill workers.

METHODS

We included workers who participated in response and cleanup activities on the water during the DWH disaster (N = 9091). PM2.5 exposures were estimated using a job-exposure matrix that linked modelled PM2.5 concentrations to detailed DWH spill work histories provided by participants. We ascertained CHD events as the first self-reported physician-diagnosed CHD or a fatal CHD event that occurred after each worker's last day of burning exposure. We estimated hazard ratios (HR) and 95% confidence intervals (95%CI) for the associations between categories of average or cumulative daily maximum PM2.5 exposure (versus a referent category of water workers not near controlled burning) and subsequent CHD. We assessed exposure-response trends by examining continuous exposure parameters in models.

RESULTS

We observed increased CHD hazard among workers with higher levels of average daily maximum exposure (low vs. referent: HR = 1.26, 95% CI: 0.93, 1.70; high vs. referent: HR = 2.11, 95% CI: 1.08, 4.12; per 10 μg/m3 increase: HR = 1.10, 95% CI: 1.02, 1.19). We also observed suggestively elevated HRs among workers with higher cumulative daily maximum exposure (low vs. referent: HR = 1.19, 95% CI: 0.68, 2.08; medium vs. referent: HR = 1.38, 95% CI: 0.88, 2.16; high vs. referent: HR = 1.44, 95% CI: 0.96, 2.14; per 100 μg/m3-d increase: HR = 1.03, 95% CI: 1.00, 1.05).

CONCLUSIONS

Among oil spill workers, exposure to PM2.5 from flaring/burning of oil/gas was associated with increased risk of CHD.

Log-transformation of Independent Variables: Must We?

Epidemiologic studies often quantify exposure using biomarkers, which commonly have statistically skewed distributions. Although normality assumption is not required if the biomarker is used as an independent variable in linear regression, it has become common practice to log-transform the biomarker concentrations. This transformation can be motivated by concerns for nonlinear dose-response relationship or outliers; however, such transformation may not always reduce bias. In this study, we evaluated the validity of motivations underlying the decision to log-transform an independent variable using simulations, considering eight scenarios that can give rise to skewed X and normal Y. Our simulation study demonstrates that (1) if the skewness of exposure did not arise from a biasing factor (e.g., measurement error), the analytic approach with the best overall model fit best reflected the underlying outcome generating methods and was least biased, regardless of the skewness of X and (2) all estimates were biased if the skewness of exposure was a consequence of a biasing factor. We additionally illustrate a process to determine whether the transformation of an independent variable is needed using NHANES. Our study and suggestion to divorce the shape of the exposure distribution from the decision to log-transform it may aid researchers in planning for analysis using biomarkers or other skewed independent variables.

Light at night and the risk of breast cancer: Findings from the Sister study

BACKGROUND

Light at night (LAN) may alter estrogen regulation through circadian disruption. High levels of outdoor LAN may increase breast cancer risk, but studies have largely not considered possible residual confounding from correlated environmental exposures. We evaluated the association between indoor and outdoor LAN and incident breast cancer.

METHODS

In 47,145 participants in the prospective Sister Study cohort living in the contiguous U.S., exposure to outdoor LAN was determined using satellite-measured residential data and indoor LAN was self-reported (light/TV on, light from outside the room, nightlight, no light). We used Cox proportional hazards models to calculate hazard ratios (HRs) and 95% confidence intervals (CIs) for the associations between outdoor and indoor LAN and breast cancer risk. Models were adjusted for age, race/ethnicity, educational attainment, annual household income, neighborhood disadvantage, latitude, and population density as a proxy for urbanicity. To evaluate the potential for residual confounding of the outdoor LAN and breast cancer relationship by factors associated with urbanicity, we considered further adjustment for exposures correlated with outdoor LAN including NO2 [Spearman correlation coefficient, rho (ρ) = 0.78], PM2.5 (ρ = 0.36), green space (ρ =  - 0.41), and noise (ρ = 0.81).

RESULTS

During 11 years of follow-up, 3,734 breast cancer cases were identified. Outdoor LAN was modestly, but non-monotonically, associated with a higher risk of breast cancer (Quintile 4 vs 1: HR = 1.10, 95% CI: 0.99-1.22; Quintile 5 vs 1: HR = 1.04, 95% CI: 0.93-1.16); however, no association was evident after adjustment for correlated ambient exposures (Quintile 4 vs 1: HR = 0.99, 95% CI: 0.86-1.14; Quintile 5 vs 1: HR = 0.89, 95% CI: 0.74-1.06). Compared to those with no indoor LAN exposure, sleeping with a light or TV on was associated with a HR = 1.09 (95% CI: 0.97-1.23) in the adjusted model.

CONCLUSIONS

Outdoor LAN does not appear to increase the risk of breast cancer after adjustment for correlated environmental exposures.

A Bespoke Instrumental Variable Approach to Correction for Exposure Measurement Error

A covariate-adjusted estimate of an exposure-outcome association may be biased if the exposure variable suffers measurement error. We propose an approach to correct for exposure measurement error in a covariate-adjusted estimate of the association between a continuous exposure variable and outcome of interest. Our proposed approach requires data for a reference population in which the exposure was a priori set to some known level (e.g., 0, and is therefore unexposed); however, our approach does not require an exposure validation study or replicate measures of exposure, which are typically needed when addressing bias due to exposure measurement error. A key condition for this method, which we refer to as "partial population exchangeability," requires that the association between a measured covariate and outcome in the reference population equals the association between that covariate and outcome in the target population in the absence of exposure. We illustrate the approach using simulations and an example.

Use of Straighteners and Other Hair Products and Incident Uterine Cancer

BACKGROUND

Hair products may contain hazardous chemicals with endocrine-disrupting and carcinogenic properties. Previous studies have found hair product use to be associated with a higher risk of hormone-sensitive cancers including breast and ovarian cancer; however, to our knowledge, no previous study has investigated the relationship with uterine cancer.

METHODS

We examined associations between hair product use and incident uterine cancer among 33 947 Sister Study participants aged 35-74 years who had a uterus at enrollment (2003-2009). In baseline questionnaires, participants in this large, racially and ethnically diverse prospective cohort self-reported their use of hair products in the prior 12 months, including hair dyes; straighteners, relaxers, or pressing products; and permanents or body waves. We estimated adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) to quantify associations between hair product use and uterine cancer using Cox proportional hazard models. All statistical tests were 2-sided.

RESULTS

Over an average of 10.9 years of follow-up, 378 uterine cancer cases were identified. Ever vs never use of straightening products in the previous 12 months was associated with higher incident uterine cancer rates (HR = 1.80, 95% CI = 1.12 to 2.88). The association was stronger when comparing frequent use (>4 times in the past 12 months) vs never use (HR = 2.55, 95% CI = 1.46 to 4.45; Ptrend = .002). Use of other hair products, including dyes and permanents or body waves, was not associated with incident uterine cancer.

CONCLUSION

These findings are the first epidemiologic evidence of association between use of straightening products and uterine cancer. More research is warranted to replicate our findings in other settings and to identify specific chemicals driving this observed association.

Comparing Trial and Real-world Adjuvant Oxaliplatin Delivery in Patients With Stage III Colon Cancer Using a Longitudinal Cumulative Dose

Importance

Delivery of adjuvant chemotherapy can differ substantially between trial and real-world populations. Adherence metrics like relative dose intensity (RDI) cannot capture the timing of modifications and mask differences in the total amount of chemotherapy received.

Objective

To compare oxaliplatin delivery between MOSAIC trial participants and patients treated in the US Oncology Network with stage III colon cancer using a longitudinal cumulative dose (LCD).

Design, Setting, and Participants

This cohort study used secondary data from the MOSAIC trial, an international randomized clinical trial (concluded in 2004), and electronic health records from US Oncology (2009-2018), a network of community oncology practices in the US. It included participants in MOSAIC with stage III colon cancer who were randomized to receive treatment with oxaliplatin and fluorouracil/leucovorin (n = 663) and US Oncology patients with stage III colon cancer who were treated with a modified FOLFOX-6 regimen (n = 2523).

Exposures

Oxaliplatin and fluorouracil/leucovorin.

Outcomes and Measures

We evaluated RDI and LCD over time and at the end of treatment in the MOSAIC and US Oncology populations. We used bootstrapping to estimate 95% confidence bands for LCD differences between the populations.

Results

The 663 MOSAIC participants (296 women [44.7%]) and 2523 US Oncology patients (1245 women [49.4%]) were generally similar with respect to demographic characteristics. Median RDI was lower in US Oncology (80% in MOSAIC vs 70% in US Oncology). The LCD also suggested differences in the total amount of oxaliplatin received between populations; the final median LCD in US Oncology was 10.2% lower than in MOSAIC, equivalent to receiving 1.2 fewer treatment cycles less of oxaliplatin. This difference only began 133 days into treatment and persisted after accounting for covariates, likely in terms of more frequent oxaliplatin treatment discontinuation in US Oncology patients than their MOSAIC counterparts.

Conclusions and Relevance

The study results suggest that real-world patients in community practice in the US treated with modified FOLFOX 6 received less oxaliplatin than their historical counterparts in the MOSAIC trial, with differences manifesting late in the treatment course. The LCD allowed us to identify the amount and extent of these differences, the timing of which was unclear when using RDI alone.

Trial Registration

ClinicalTrials.gov identifier: NCT00275210.

Associations between airborne crude oil chemicals and symptom-based asthma

RATIONALE

The 2010 Deepwater Horizon (DWH) oil spill response and cleanup (OSRC) workers were exposed to airborne total hydrocarbons (THC), benzene, toluene, ethylbenzene, o-, m-, and p-xylenes and n-hexane (BTEX-H) from crude oil and PM2.5 from burning/flaring oil and natural gas. Little is known about asthma risk among oil spill cleanup workers.

OBJECTIVES

We assessed the relationship between asthma and several oil spill-related exposures including job classes, THC, individual BTEX-H chemicals, the BTEX-H mixture, and PM2.5 using data from the Gulf Long-Term Follow-up (GuLF) Study, a prospective cohort of 24,937 cleanup workers and 7,671 nonworkers following the DWH disaster.

METHODS

Our analysis largely focused on the 19,018 workers without asthma before the spill who had complete exposure, outcome, and covariate information. We defined incident asthma 1-3 years following exposure using both self-reported wheeze and self-reported physician diagnosis of asthma. THC and BTEX-H were assigned to participants based on measurement data and work histories, while PM2.5 used modeled estimates. We used modified Poisson regression to estimate risk ratios (RR) and 95% confidence intervals (CIs) for associations between spill-related exposures and asthma and a quantile-based g-computation approach to explore the joint effect of the BTEX-H mixture on asthma risk.

RESULTS

OSRC workers had greater asthma risk than nonworkers (RR: 1.60, 95% CI: 1.38, 1.85). Higher estimated THC exposure levels were associated with increased risk in an exposure-dependent manner (linear trend test p < 0.0001). Asthma risk also increased with increasing exposure to individual BTEX-H chemicals and the chemical mixture: A simultaneous quartile increase in the BTEX-H mixture was associated with an increased asthma risk of 1.45 (95% CI: 1.35,1.55). With fewer cases, associations were less apparent for physician-diagnosed asthma alone.

CONCLUSIONS

THC and BTEX-H were associated with increased asthma risk defined using wheeze symptoms as well as a physician diagnosis.

Associations between mixtures of urinary phthalate metabolite concentrations and oxidative stress biomarkers among couples undergoing fertility treatment

Phthalate exposure has been associated with adverse reproductive outcomes and oxidative stress is a potential mechanism by which they act. However, few human studies have explored co-exposure confounding or joint effects. Furthermore, most studies examine associations between biomarkers of exposure and oxidative stress from the same urine sample. We investigated single-exposure, co-exposure-adjusted, and joint associations between phthalate metabolites and oxidative stress in the Environment and Reproductive Health (EARTH) study among couples undergoing fertility treatment. We examined cross-sectional associations in both women and men, and longitudinal associations in women. Urine was collected in the follicular phase (women only) and at the time of fertility procedure (women and men), and analyzed for 11 phthalate metabolites. Urine from the time of fertility procedure was analyzed for oxidative stress biomarkers, including free 8-iso-prostaglandin F2α (8-iso-PGF2α), its primary metabolite (2,3-dinor-5,6-dihydro-15-F2t-isoprostane [F2-IsoP-M]), and prostaglandin F2α (PGF2α). Linear mixed effects models were used to estimate single-exposure associations. Bayesian Kernel Machine Regression (BKMR) was used to adjust for co-exposures and to estimate joint effects. Among women, we observed positive associations between all phthalate metabolites and oxidative stress biomarkers in single-exposure models, but there was clear co-exposure confounding. For instance, in a single-exposure model, we estimated a 63% (95% confidence interval: 51, 77) increase in the 8-iso-PGF2α metabolite per interquartile range (IQR) difference in mono-n-butyl phthalate (MBP) versus a 34% (95% credible interval: 12, 60) increase in co-adjusted models. However, several phthalate metabolites remained associated with oxidative stress in co-exposure models, and the joint effects of all exposures were high (e.g., an 114% increase in the 8-iso-PGF2α metabolite per IQR difference in all exposures). Longitudinal results were also attenuated compared to cross-sectional results in women; however, the joint effect of all exposures and the 8-iso-PGF2α metabolite remained positive and statistically significant (11% increase per IQR difference in all exposures, 95% credible interval: 0.2, 23). In men, associations were generally less pronounced, although the joint effect of the mixture on 8-iso-PGF2α was above the null. Because oxidative stress is related to reproductive success among couples seeking fertility treatment, mitigating phthalate exposure should be considered as a potentially beneficial measure.

Associations Between Prenatal Urinary Biomarkers of Phthalate Exposure and Preterm Birth: A Pooled Study of 16 US Cohorts

Importance

Phthalate exposure is widespread among pregnant women and may be a risk factor for preterm birth.

Objective

To investigate the prospective association between urinary biomarkers of phthalates in pregnancy and preterm birth among individuals living in the US.

Design, Setting, and Participants

Individual-level data were pooled from 16 preconception and pregnancy studies conducted in the US. Pregnant individuals who delivered between 1983 and 2018 and provided 1 or more urine samples during pregnancy were included.

Exposures

Urinary phthalate metabolites were quantified as biomarkers of phthalate exposure. Concentrations of 11 phthalate metabolites were standardized for urine dilution and mean repeated measurements across pregnancy were calculated.

Main Outcomes and Measures

Logistic regression models were used to examine the association between each phthalate metabolite with the odds of preterm birth, defined as less than 37 weeks of gestation at delivery (n = 539). Models pooled data using fixed effects and adjusted for maternal age, race and ethnicity, education, and prepregnancy body mass index. The association between the overall mixture of phthalate metabolites and preterm birth was also examined with logistic regression. G-computation, which requires certain assumptions to be considered causal, was used to estimate the association with hypothetical interventions to reduce the mixture concentrations on preterm birth.

Results

The final analytic sample included 6045 participants (mean [SD] age, 29.1 [6.1] years). Overall, 802 individuals (13.3%) were Black, 2323 (38.4%) were Hispanic/Latina, 2576 (42.6%) were White, and 328 (5.4%) had other race and ethnicity (including American Indian/Alaskan Native, Native Hawaiian, >1 racial identity, or reported as other). Most phthalate metabolites were detected in more than 96% of participants. Higher odds of preterm birth, ranging from 12% to 16%, were observed in association with an interquartile range increase in urinary concentrations of mono-n-butyl phthalate (odds ratio [OR], 1.12 [95% CI, 0.98-1.27]), mono-isobutyl phthalate (OR, 1.16 [95% CI, 1.00-1.34]), mono(2-ethyl-5-carboxypentyl) phthalate (OR, 1.16 [95% CI, 1.00-1.34]), and mono(3-carboxypropyl) phthalate (OR, 1.14 [95% CI, 1.01-1.29]). Among approximately 90 preterm births per 1000 live births in this study population, hypothetical interventions to reduce the mixture of phthalate metabolite levels by 10%, 30%, and 50% were estimated to prevent 1.8 (95% CI, 0.5-3.1), 5.9 (95% CI, 1.7-9.9), and 11.1 (95% CI, 3.6-18.3) preterm births, respectively.

Conclusions and Relevance

Results from this large US study population suggest that phthalate exposure during pregnancy may be a preventable risk factor for preterm delivery.

Innovations in suicide prevention research (INSPIRE): a protocol for a population-based case-control study

BACKGROUND

Suicide deaths have been increasing for the past 20 years in the USA resulting in 45 979 deaths in 2020, a 29% increase since 1999. Lack of data linkage between entities with potential to implement large suicide prevention initiatives (health insurers, health institutions and corrections) is a barrier to developing an integrated framework for suicide prevention.

OBJECTIVES

Data linkage between death records and several large administrative datasets to (1) estimate associations between risk factors and suicide outcomes, (2) develop predictive algorithms and (3) establish long-term data linkage workflow to ensure ongoing suicide surveillance.

METHODS

We will combine six data sources from North Carolina, the 10th most populous state in the USA, from 2006 onward, including death certificate records, violent deaths reporting system, large private health insurance claims data, Medicaid claims data, University of North Carolina electronic health records and data on justice involved individuals released from incarceration. We will determine the incidence of death from suicide, suicide attempts and ideation in the four subpopulations to establish benchmarks. We will use a nested case-control design with incidence density-matched population-based controls to (1) identify short-term and long-term risk factors associated with suicide attempts and mortality and (2) develop machine learning-based predictive algorithms to identify individuals at risk of suicide deaths.

DISCUSSION

We will address gaps from prior studies by establishing an in-depth linked suicide surveillance system integrating multiple large, comprehensive databases that permit establishment of benchmarks, identification of predictors, evaluation of prevention efforts and establishment of long-term surveillance workflow protocols.

Prenatal PFAS and psychosocial stress exposures in relation to fetal growth in two pregnancy cohorts: Applying environmental mixture methods to chemical and non-chemical stressors

BACKGROUND

Prenatal exposure to individual per‑ and poly‑fluoroalkyl substances (PFAS) and psychosocial stressors have been associated with reductions in fetal growth. Studies suggest cumulative or joint effects of chemical and non-chemical stressors on fetal growth. However, few studies have examined PFAS and non-chemical stressors together as a mixture, which better reflects real life exposure patterns. We examined joint associations between PFAS, perceived stress, and depression, and fetal growth using two approaches developed for exposure mixtures.

METHODS

Pregnant participants were enrolled in the Chemicals in Our Bodies cohort and Illinois Kids Development Study, which together make up the ECHO.CA.IL cohort. Seven PFAS were previously measured in 2nd trimester maternal serum samples and were natural log transformed for analyses. Perceived stress and depression were assessed using self-reported validated questionnaires, which were converted to t-scores using validated methods. Quantile g-computation and Bayesian kernel machine regression (BKMR) were used to assess joint associations between PFAS, perceived stress and depression t-scores and birthweight z-scores (N = 876).

RESULTS

Individual PFAS, depression and perceived stress t-scores were negatively correlated with birthweight z-scores. Using quantile g-computation, a simultaneous one quartile increase in all PFAS, perceived stress and depression t-scores was associated with a slight reduction in birthweight z-scores (mean change per quartile increase = -0.09, 95% confidence interval = -0.21,0.03). BKMR similarly indicated that cumulative PFAS and stress t-scores were modestly associated with lower birthweight z-scores. Across both methods, the joint association appeared to be distributed across multiple exposures rather than due to a single exposure.

CONCLUSIONS

Our study is one of the first to examine the joint effects of chemical and non-chemical stressors on fetal growth using mixture methods. We found that PFAS, perceived stress, and depression in combination were modestly associated were lower birthweight z-scores, which supports prior studies indicating that chemical and non-chemical stressors are jointly associated with adverse health outcomes.

Urinary phthalate metabolite mixtures in pregnancy and fetal growth: Findings from the infant development and the environment study

BACKGROUND

Prenatal phthalate exposure has been linked to reductions in fetal growth in animal and laboratory studies, but epidemiologic evidence is equivocal.

OBJECTIVE

Examine the association between prenatal phthalate metabolite mixtures and fetal growth and evaluate whether that association is modified by fetal sex or omega-3 intake during pregnancy.

METHODS

Analyses included 604 singleton pregnancies from TIDES, a prospective pregnancy cohort with spot urine samples and questionnaires collected in each trimester. Pregnancy-averaged phthalate exposure estimates were calculated as the geometric means of specific-gravity corrected phthalate metabolites. Fetal growth outcomes included birthweight and length, and ultrasound-derived size and velocity of estimated fetal weight, femur length, abdominal and head circumferences in the second and third trimesters. We used a novel application of quantile g-computation to estimate the joint association between pregnancy-averaged phthalate exposure and fetal growth, and to examine effect modification of that association by infant sex or omega-3 intake during pregnancy.

RESULTS

There were few statistically significant differences in birth size and fetal growth by exposure. A one-quartile increase in the phthalate mixture was modestly associated with reduced birthweight(β [95% confidence interval)]: -54.6 [-128.9, 19.7] grams; p = 0.15) and length (-0.2 [-0.6, 0.2] centimeters; p = 0.40). A one-quartile increase in the phthalate mixture was associated with reduced birth length in males (-0.5 [-1.0, 0.0] centimeters) but not for females (0.1 [-0.2, 0.3] centimeters); interaction p = 0.05. The phthalate metabolite mixture was inversely associated with ultrasound-derived fetal growth among those with adequate omega-3 intake. For example, a one-quartile increase in the phthalate mixture was associated with reduced abdominal circumference in the third trimesters in those with adequate omega-3 intake (-3.3 [-6.8, 0.1] millimeters) but not those with inadequate omega-3 intake (1.8 [-0.8, 4.5] millimeters); interaction p = 0.01.

CONCLUSION

Prenatal phthalate exposure was not significantly associated with fetal growth outcomes, with some exceptions for certain subgroups.

The Case for Case-Cohort: An Applied Epidemiologist's Guide to Reframing Case-Cohort Studies to Improve Usability and Flexibility

When research questions require the use of precious samples, expensive assays or equipment, or labor-intensive data collection or analysis, nested case-control or case-cohort sampling of observational cohort study participants can often reduce costs. These study designs have similar statistical precision for addressing a singular research question, but case-cohort studies have broader efficiency and superior flexibility. Despite this, case-cohort designs are comparatively underutilized in the epidemiologic literature. Recent advances in statistical methods and software have made analyses of case-cohort data easier to implement, and advances from casual inference, such as inverse probability of sampling weights, have allowed the case-cohort design to be used with a variety of target parameters and populations. To provide an accessible link to this technical literature, we give a conceptual overview of case-cohort study analysis with inverse probability of sampling weights. We show how this general analytic approach can be leveraged to more efficiently study subgroups of interest or disease subtypes or to examine associations independent of case status. A brief discussion of how this framework could be extended to incorporate other related methodologic applications further demonstrates the broad cost-effectiveness and adaptability of case-cohort methods for a variety of modern epidemiologic applications in resource-limited settings.

Trends in fatal occupational injuries in Latino/a workers relative to other groups, North Carolina 2000-2017

BACKGROUND

Latino/a workers may experience higher fatal occupational injury rates than non-Latino/a workers. In North Carolina, the Latino/a population more than doubled between 2000 and 2017. We examined fatal occupational injuries among Latino/a and non-Latino/a workers in North Carolina over this period.

METHODS

Information on fatal occupational injuries was abstracted from records of the North Carolina Office of the Chief Medical Examiner and the death certificate records held by the North Carolina Office of Vital Records. Estimates of the working population were derived from the decennial census and American Community Survey. Estimates of annual rates of fatal occupational injury for the period January 1, 2000 to December 31, 2017 were derived for Latino/a workers and compared to Black and White workers not identified as Latino/a.

RESULTS

Over the study period, 1,783 fatal occupational injuries were identified among non-Latino/a workers and 259 fatal occupational injuries among Latino/a workers in North Carolina. The majority of fatal occupational injuries among Latino/a workers occurred among males employed in construction and agriculture. While the fatal occupational injury rate among Latino/a workers declined over the study period, the rate among Latino/a workers was higher than among non-Latino/a White and Black workers; moreover, fatal occupational injury rates for Latino/a workers trended upwards during the most recent years of the study period.

CONCLUSIONS

Latino/a workers in North Carolina have the highest fatal occupational injury rate of any race/ethnicity group.

Real and Illusory Precision Gains in Meta-Analysis to Speed Action on Carcinogens

One should avoid benzene exposure, all other things being equal. Risk assessment can help inform human health outcomes when all other things are not equal, as when competing legal or economic interests arise. In sparse literatures where exposures may be highly deleterious and yet understudied, there is a dire need for evidence synthesis, such as meta-analysis, to maximally inform risk assessment. Here, using the analysis and approach of Scholten and colleagues from the current issue as a touch point, I describe how meta-analysis could ideally meet this aim and how it often fails to do so. Some of the current literature on transportability of causal effects is illustrative, and I describe how some of the lessons from this literature could be applied within the innovative framework of Scholten and colleagues to leverage meta-analysis within the broader decision-making framework of risk-assessment. See related article by Scholten et al., p. 751.

Prenatal Phthalate Exposure and Child Weight and Adiposity from in Utero to 6 Years of Age

BACKGROUND

Prenatal phthalate exposure has been associated with lower birth weight but also higher weight in childhood. Few studies have examined weight or adiposity from birth to childhood and thus cannot assess growth trajectories associated with exposure.

OBJECTIVE

We assessed associations between maternal phthalate exposures in pregnancy and child weight and adiposity measured prenatally through childhood (3-6 years of age).

METHODS

Within The Infant Development and the Environment Study (TIDES), a prospective pregnancy cohort, we analyzed a panel of phthalate metabolites in urine collected at two visits from early and late gestation (N=780). We estimated average phthalate metabolite associations with child weight z-scores from ∼20wk gestation (estimated by ultrasound), birth, and 1, 3, 4, and 6 years of age using linear mixed-effects (LME) models. We also modeled associations with adiposity z-scores from birth (weight for length) and 1, 3, 4, and 6 years of age [body mass index (BMI)] using LME models.

RESULTS

For weight, we observed inverse associations between several phthalate metabolites and birth weight z-scores, but no associations were observed with postnatal weight z-scores in LME models. Regarding adiposity, we observed inverse associations between phthalate metabolites and weight-for-length z-scores at birth, but positive associations were observed with BMI z-scores at 3-4 years of age in LME models. For example, mono-ethyl phthalate was associated with a 0.17-unit decrease in birth weight-for-length z-score [95% confidence interval (CI): -0.29, -0.05] and a 0.18-unit increase in 4-years-of-age BMI z-score (95% CI: 0.04, 0.32).

DISCUSSION

We observed associations between prenatal exposure to phthalates and lower weight at birth but not at childhood follow-up visits. However, for adiposity, we observed an interesting pattern of association with low adiposity at delivery as well as high adiposity at 3-4 years of age. Although it is not clear from our results whether these associations occur within the same children, such a pattern of adiposity in early life has been linked to cardiometabolic disease in adulthood and deserves special attention as an outcome in the study of prenatal exposures in the developmental origins of health and disease. https://doi.org/10.1289/EHP10077.

Comparison of Approaches for Measuring Adherence and Persistence to Oral Oncologic Therapies in Patients Diagnosed with Metastatic Renal Cell Carcinoma

BACKGROUND

Adherence and persistence studies face several methodologic difficulties, including short-term mortality. We compared approaches to quantify adherence and persistence to first line (1L) oral targeted therapy (TT) in patients diagnosed with metastatic renal cell carcinoma (mRCC).

METHODS

Patients with mRCC ages 66 years or more who initiated TTs within 4 months of diagnosis were identified in the Surveillance, Epidemiology, and End Results Medicare-linked database (2007-2015). Adherence [proportion of days covered (PDC) >80%] was calculated using (i) PDC with a fixed 6-month denominator including then excluding patients who died within the 6 months and (ii) PDC with a denominator measuring time on treatment. Risk of nonpersistence was obtained by censoring death or treating death as a competing risk using cumulative incidence functions.

RESULTS

Among 485 patients with mRCC initiating a 1L oral TT (sunitinib, 64%; pazopanib, 25%; other, 11%), 40% died within 6 months. Adherence was higher after restricting to patients who survived (60%) compared with including those patients and assigning zero days covered after death (47%). Risk of nonpersistence was higher when censoring patients at death, 0.91 [95% confidence interval (CI), 0.88-0.94], compared with treating death as a competing risk, 0.75 (95% CI, 0.71-0.79).

CONCLUSIONS

Different approaches to handling death resulted in different adherence and persistence estimates in the metastatic setting. Future studies should explicitly report the proportion of patient deaths over time and explore appropriate methods to account for death as competing risk.

IMPACT

Use of several approaches can provide a more comprehensive picture of medication-taking behavior in the metastatic setting where death is a major competing risk.

Analysis of the novel NCWELL database highlights two decades of co-occurrence of toxic metals in North Carolina private well water: Public health and environmental justice implications

Private well users are particularly vulnerable to metal exposure as they are not protected by the Safe Drinking Water Act. In North Carolina (NC), approximately 2.4 million individuals rely on private well water. In the present study, we constructed the NCWELL database: a comprehensive database of 117,960 geocoded well water tests over twenty-years in NC inclusive of 28 metals/metalloids. The NCWELL database was analyzed to identify areas of concern for single and co-occurring toxic metal contamination of private wells in NC. County-level population-at-risk rankings were calculated by combining toxic metal levels and the proportion of residents relying on well water. Additionally, k-means analysis was used to identify counties with critical co-occurrence of toxic metals. In the NCWELL database, inorganic arsenic (iAs) and lead (Pb) were detected above the EPA standards of 10 and 15 ppb in over 2500 and over 3000 tests, respectively. Shockingly, iAs was observed at levels up to 806 ppb and Pb at levels up to 105,440 ppb. Manganese (Mn) was detected above the EPA lifetime Health Advisory Limit in 4.9% and above the secondary Maximum Contaminant Level in 24.3% of all well water tests in NC, with a maximum concentration of 46,300 ppb reported. Mixtures-based analysis identified four distinct clusters of counties, one demonstrating high iAs and Mn and another with high Pb. Over the twenty-year period, metal levels remained high, indicative of sustained contamination in areas of concern. This study provides a novel database for researchers and concerned citizens in NC, demonstrates a methodology for identifying priority geographic regions for single and multiple contaminants, and has environmental justice implications in NC where metal exposure via private well water remains a serious public health concern.

Life expectancy in older adults with advanced cancer: Evaluation of a geriatric assessment-based prognostic model

OBJECTIVES

Oncologists estimate patients' prognosis to guide care. Evidence suggests oncologists tend to overestimate life expectancy, which can lead to care with questionable benefits. Information obtained from geriatric assessment may improve prognostication for older adults. In this study, we created a geriatric assessment-based prognostic model for older adults with advanced cancer and compared its performance to alternative models.

MATERIALS AND METHODS

We conducted a secondary analysis of a trial (URCC 13070; PI: Mohile) capturing geriatric assessment and vital status up to one year for adults age ≥ 70 years with advanced cancer. Oncologists estimated life expectancy as 0-6 months, 7-12 months, and > 1 year. Three statistical models were developed: (1) a model including age, sex, cancer type, and stage (basic model), (2) basic model + Karnofsky Performance Status (≤50, 60-70, and 80+) (KPS model), and (3) basic model +16 binary indicators of geriatric assessment impairments (GA model). Cox regression was used to model one-year survival; c-indices and time-dependent c-statistics assessed model discrimination and stratified survival curves assessed model calibration.

RESULTS

We included 484 participants; mean age was 75; 48% had gastrointestinal or lung cancer. Overall, 43% of patients died within one year. Oncologists classified prognosis accurately for 55% of patients, overestimated for 35%, and underestimated for 10%. C-indices were 0.61 (basic model), 0.62 (KPS model), and 0.63 (GA model). The GA model was well-calibrated.

CONCLUSIONS

The GA model showed moderate discrimination for survival, similar to alternative models, but calibration was improved. Further research is needed to optimize geriatric assessment-based prognostic models for use in older adults with advanced cancer.

Amplification of Bias Due to Exposure Measurement Error

Observational epidemiologic studies typically face challenges of exposure measurement error and confounding. Consider an observational study of the association between a continuous exposure and an outcome, where the exposure variable of primary interest suffers from classical measurement error (i.e., the measured exposures are distributed around the true exposure with independent error). In the absence of exposure measurement error, it is widely recognized that one should control for confounders of the association of interest to obtain an unbiased estimate of the effect of that exposure on the outcome of interest. However, here we show that, in the presence of classical exposure measurement error, the net bias in an estimate of the association of interest may increase upon adjustment for confounders. We offer an analytical expression for calculating the change in net bias in an estimate of the association of interest upon adjustment for a confounder in the presence of classical exposure measurement error, and we illustrate this problem using simulations.

Vitamin D Supplement Use and Risk of Breast Cancer by Race-Ethnicity

BACKGROUND

Vitamin D has anticarcinogenic properties, but a relationship between vitamin D supplement use and breast cancer is not established. Few studies have accounted for changes in supplement use over time or evaluated racial-ethnic differences.

METHODS

The Sister Study is a prospective cohort of 50,884 women with 35-74 years of age who had a sister with breast cancer, but no breast cancer themselves at enrollment (2003-2009). We used Cox proportional hazards models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for the association between vitamin D supplement use and incident breast cancer (3,502 cases; median follow-up 10.5 years).

RESULTS

Vitamin D supplement use was common, with 64% reporting ever use (at least once per month) in the year before enrollment. Considering supplement use over time, ever use of vitamin D supplements was not meaningfully associated with breast cancer (HR = 0.96, 95% CI = 0.88, 1.0), relative to never use. However, after adjusting for prior use, recent use of vitamin D supplements ≥1/month was inversely associated with breast cancer (HR = 0.88, 95% CI = 0.78, 1.0), relative to nonrecent use. The inverse association was stronger for ductal carcinoma in situ (HR = 0.67, 95% CI = 0.52, 0.87) than invasive breast cancer (HR = 0.94, 95% CI = 0.72, 1.1, p-for-heterogeneity = 0.02). Supplement use was less common among African American/Black (56%) and non-Black Hispanic/Latina (50%) women than non-Hispanic White women (66%), but there was limited evidence of racial-ethnic differences in HRs (p-for-heterogeneity = 0.16 for ever use, P = 0.55 for recent).

CONCLUSIONS

Our findings are consistent with the hypothesis that recent vitamin D use is inversely associated with breast cancer risk.

Prenatal Exposure to Nonpersistent Chemical Mixtures and Offspring IQ and Emotional and Behavioral Problems

Prenatal exposure to nonpersistent chemicals such as phthalates, bisphenols, and organophosphate (OP) pesticides is ubiquitous and occurs in mixtures. So far, epidemiological studies investigating neurodevelopmental consequences of these exposures have mainly been restricted to single-pollutant models. Thus, we studied the association between prenatal exposure to nonpersistent chemical mixtures and child IQ and emotional and behavioral problems. Data came from 782 mother-child pairs. Eleven phthalate, one bisphenol, and five OP pesticide urinary exposure biomarkers were measured three times during pregnancy and averaged. Nonverbal IQ, internalizing and attention problems, aggressive behavior, and autistic traits were assessed at child age 6 years. We used quantile g-computation to estimate the change in each outcome per quartile increase in all chemicals within the mixture. Higher exposure to the mixture was associated with lower nonverbal IQ (-4.0 points (95%CI = -7.0, -1.0), -5.5 points (95%CI = -10.2, -0.9), and -4.6 points (95%CI = -10.8, 1.5) for the second, third, and fourth quartile, respectively, compared to the first quartile). These results were mainly driven by the phthalate mixture. No association was observed with emotional and behavioral problems. Prenatal exposure to nonpersistent chemical mixtures was associated with lower nonverbal IQ in children. Exposure to chemical mixtures during gestation is universal and may impact neurodevelopment.

Bayesian G-Computation for Estimating Impacts of Interventions on Exposure Mixtures: Demonstration With Metals From Coal-Fired Power Plants and Birth Weight

The importance of studying the health impacts of exposure mixtures is increasingly being recognized, but such research presents many methodological and interpretation difficulties. We used Bayesian g-computation to estimate effects of a simulated public health action on exposure mixtures and birth weights in Milwaukee, Wisconsin, in 2011-2013. We linked data from birth records with census-tract-level air toxics data from the Environmental Protection Agency's National Air Toxics Assessment model. We estimated the difference between observed and expected birth weights that theoretically would have followed a hypothetical intervention to reduce exposure to 6 airborne metals by decommissioning 3 coal-fired power plants in Milwaukee County prior to 2010. Using Bayesian g-computation, we estimated a 68-g (95% credible interval: 25, 135) increase in birth weight following this hypothetical intervention. This example demonstrates the utility of our approach for using observational data to evaluate and contrast possible public health actions. Additionally, Bayesian g-computation offers a flexible strategy for estimating the effects of highly correlated exposures, addressing statistical issues such as variance inflation, and addressing conceptual issues such as the lack of interpretability of independent effects.