Dataset Information

A Quantile-Based g-Computation Approach to Addressing the Effects of Exposure Mixtures.

ABSTRACT:

Background

Exposure mixtures frequently occur in data across many domains, particularly in the fields of environmental and nutritional epidemiology. Various strategies have arisen to answer questions about exposure mixtures, including methods such as weighted quantile sum (WQS) regression that estimate a joint effect of the mixture components.

Objectives

We demonstrate a new approach to estimating the joint effects of a mixture: quantile g-computation. This approach combines the inferential simplicity of WQS regression with the flexibility of g-computation, a method of causal effect estimation. We use simulations to examine whether quantile g-computation and WQS regression can accurately and precisely estimate the effects of mixtures in a variety of common scenarios.

Methods

We examine the bias, confidence interval (CI) coverage, and bias-variance tradeoff of quantile g-computation and WQS regression and how these quantities are impacted by the presence of noncausal exposures, exposure correlation, unmeasured confounding, and nonlinearity of exposure effects.

Results

Quantile g-computation, unlike WQS regression, allows inference on mixture effects that is unbiased with appropriate CI coverage at sample sizes typically encountered in epidemiologic studies and when the assumptions of WQS regression are not met. Further, WQS regression can magnify bias from unmeasured confounding that might occur if important components of the mixture are omitted from the analysis.

Discussion

Unlike inferential approaches that examine the effects of individual exposures while holding other exposures constant, methods like quantile g-computation that can estimate the effect of a mixture are essential for understanding the effects of potential public health actions that act on exposure sources. Our approach may serve to help bridge gaps between epidemiologic analysis and interventions such as regulations on industrial emissions or mining processes, dietary changes, or consumer behavioral changes that act on multiple exposures simultaneously. https://doi.org/10.1289/EHP5838.

SUBMITTER: Keil AP

PROVIDER: S-EPMC7228100 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundMolar sums are often used to quantify total phthalate exposure, but they do not capture patterns of exposure to multiple phthalates.ObjectiveTo introduce an exposure burden score method for quantifying exposure to phthalate metabolites and examine the association between phthalate burden scores and Homeostatic Model Assessment for Insulin Resistance (HOMA-IR).MethodsWe applied item response theory (IRT) to data from 3474 adults aged 20-60 years in the 2013-2018 National Health and Examination Survey (NHANES) to quantify latent phthalate exposure burden from 12 phthalate metabolites. We compared model fits of three IRT models that used different a priori groupings (general phthalate burden; low molecular weight (LMW) and high molecular weight (HMW) burdens; and LMW, HMW and DEHP burden), and used the best fitting model to estimate phthalate exposure burden scores. Regression models assessed the covariate-adjusted association between phthalate burden scores and HOMA-IR. We compared findings to those using molar sums. In secondary analyses, we examined how the IRT model could be used for data harmonization when a subset of participants are missing some phthalate metabolites, and accounted for measurement error of the phthalate burden scores in estimating associations with HOMA-IR through a resampling approach using plausible value imputation.ResultsA three correlated factors model (LMW, HMW and DEHP burdens) provided the best fit. One interquartile range (IQR) increase in DEHP burden score was associated with 0.094 (95% CI: 0.022, 0.164, p = 0.010) increase in log HOMA-IR, co-adjusted for LMW and HMW burden scores. Findings were consistent when using log molar sums. Associations of phthalate burden and insulin resistance were also consistent when participants were simulated to be missing some phthalate metabolites, and when we accounted for measurement error in estimating burden scores.ConclusionBoth phthalate molar sums and burden scores are sensitive to associations with insulin resistance. Phthalate burden scores may be useful for data harmonization.

Project description:Individuals are exposed to a large number of diverse environmental chemicals simultaneously and the evaluation of multiple chemical exposures is important for identifying cancer risk factors. The measurement of a large number of chemicals (the exposome) in epidemiologic studies is allowing for a more comprehensive assessment of cancer risk factors than was done in earlier studies that focused on only a few chemicals. Empirical evidence from epidemiologic studies shows that chemicals from different chemical classes have different magnitudes and directions of association with cancers. Given increasing data availability, there is a need for the development and assessment of statistical methods to model environmental cancer risk that considers a large number of diverse chemicals with different effects for different chemical classes. The method of grouped weighted quantile sum (GWQS) regression allows for multiple groups of chemicals to be considered in the model such that different magnitudes and directions of associations are possible for each group of chemicals. In this paper, we assessed the ability of GWQS regression to estimate exposure effects for multiple chemical groups and correctly identify important chemicals in each group using a simulation study. We compared the performance of GWQS regression with WQS regression, the least absolute shrinkage and selection operator (lasso), and the group lasso in estimating exposure effects and identifying important chemicals. The simulation study results demonstrate that GWQS is an effective method for modeling exposure to multiple groups of chemicals and compares favorably with other methods used in mixture analysis. As an application, we used GWQS regression in the California Childhood Leukemia Study (CCLS), a population-based case-control study of childhood leukemia in California to estimate exposure effects for many chemical classes while also adjusting for demographic factors. The CCLS analysis found evidence of a positive association between exposure to the herbicide dacthal and an increased risk of childhood leukemia.

Project description:Exposure to natural food contaminants during infancy may influence health consequences later in life. Hence, breast milk may serve as a vehicle to transport these contaminants, including mycotoxins, from mothers to their infants. Analytical methods mostly focused on single exposures in the past, thus neglecting co-occurrences and mixture effects. Here, we present a highly sensitive multi-biomarker approach by a sophisticated combination of steps during sample preparation including a Quick, Easy, Cheap, Effective, Rugged and Safe (QuEChERS) extraction followed by a solid phase extraction (SPE) cleanup and utilizing stable isotopes for compensating challenging matrix effects. The assay was validated in-house, reaching limits of detection (LOD) for all 34 analytes in the range of 0.1 to 300 ng/L with satisfying extraction efficiencies (75-109%) and stable intermediate precisions (1-18%) for most analytes. Compared to a similar multi-mycotoxin assay for breast milk, LOD values were decreased by a factor of 2-60x enabling the assessment of chronic low-dose exposures. The new method was applied to a small set of Nigerian breast milk samples (n = 3) to compare results with already published data. Concentration levels of samples that were found to be contaminated before could be confirmed. In addition, other mycotoxins were determined in all three samples, for example the newly investigated alternariol monomethyl ether (AME) was found for the first time in this biological fluid at concentrations up to 25 ng/L. Moreover, in a pooled Austrian sample obtained from a milk bank, trace amounts of multiple mycotoxins including AME (1.9 ng/L), beauvericin (5.4 ng/L), enniatin B (4.7 ng/L), enniatin B1 (<LOQ), ochratoxin A (<LOQ) and the estrogenic zearalenone (<LOQ) confirmed co-occurrence and exposure even in a country with high food safety standards. In conclusion, the method facilitates the determination of mycotoxins at ultra-trace levels in breast milk, enabling the generation of occurrence data necessary for comprehensive co-exposure assessment.