1
|
Hao W, Cathey AL, Aung MM, Boss J, Meeker JD, Mukherjee B. Statistical methods for chemical mixtures: a roadmap for practitioners. MEDRXIV : THE PREPRINT SERVER FOR HEALTH SCIENCES 2024:2024.03.03.24303677. [PMID: 38496435 PMCID: PMC10942527 DOI: 10.1101/2024.03.03.24303677] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 03/19/2024]
Abstract
Quantitative characterization of the health impacts associated with exposure to chemical mixtures has received considerable attention in current environmental and epidemiological studies. With many existing statistical methods and emerging approaches, it is important for practitioners to understand when each method is best suited for their inferential goals. In this study, we conduct a review and comparison of 11 analytical methods available for use in mixtures research, through extensive simulation studies for continuous and binary outcomes. These methods fall in three different classes: identifying important components of a mixture, identifying interactions and creating a summary score for risk stratification and prediction. We carry out an illustrative data analysis in the PROTECT birth cohort from Puerto Rico. Most importantly we develop an integrated package "CompMix" that provides a platform for mixtures analysis where the practitioner can implement a pipeline for several types of mixtures analysis. Our simulation results suggest that the choice of methods depends on the goal of analysis and there is no clear winner across the board. For selection of important toxicants in the mixture and for identifying interactions, Elastic net by Zou et al. (Enet), Lasso for Hierarchical Interactions by Bien et al (HierNet), Selection of nonlinear interactions by a forward stepwise algorithm by Narisetty et al. (SNIF) have the most stable performance across simulation settings. Additionally, the predictive performance of the Super Learner ensembling method by Van de Laan et al. and HierNet are found to be superior to the rest of the methods. For overall summary or a cumulative measure, we find that using the Super Learner to combine multiple Environmental Risk Scores can lead to improved risk stratification properties. We have developed an R package "CompMix: A comprehensive toolkit for environmental mixtures analysis", allowing users to implement a variety of tasks under different settings and compare the findings. In summary, our study offers guidelines for selecting appropriate statistical methods for addressing specific scientific questions related to mixtures research. We identify critical gaps where new and better methods are needed.
Collapse
|
2
|
Ryva BA, Pacyga DC, Anderson KY, Calafat AM, Whalen J, Aung MT, Gardiner JC, Braun JM, Schantz SL, Strakovsky RS. Associations of urinary non-persistent endocrine disrupting chemical biomarkers with early-to-mid pregnancy plasma sex-steroid and thyroid hormones. ENVIRONMENT INTERNATIONAL 2024; 183:108433. [PMID: 38219543 PMCID: PMC10858740 DOI: 10.1016/j.envint.2024.108433] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 09/01/2023] [Revised: 11/22/2023] [Accepted: 01/06/2024] [Indexed: 01/16/2024]
Abstract
BACKGROUND/OBJECTIVES Pregnant women are exposed to numerous endocrine disrupting chemicals (EDCs) that can affect hormonal pathways regulating pregnancy outcomes and fetal development. Thus, we evaluated overall and fetal sex-specific associations of phthalate/replacement, paraben, and phenol biomarkers with sex-steroid and thyroid hormones. METHODS Illinois women (n = 302) provided plasma for progesterone, estradiol, testosterone, free T4 (FT4), total T4 (TT4), and thyroid stimulating hormone (TSH) at median 17 weeks gestation. Women also provided up-to-five first-morning urine samples monthly across pregnancy (8-40 weeks), which we pooled to measure 19 phthalate/replacement metabolites (reflecting ten parent compounds), three parabens, and six phenols. We used linear regression to evaluate overall and fetal sex-specific associations of biomarkers with hormones, as well as weighted quantile sum and Bayesian kernel machine regression (BKMR) to assess cumulative associations, non-linearities, and chemical interactions. RESULTS In women of relatively high socioeconomic status, several EDC biomarkers were associated with select hormones, without cumulative or non-linear associations with progesterone, FT4, or TT4. The biomarker mixture was negatively associated with estradiol (only at higher biomarker concentrations using BKMR), testosterone, and TSH, where each 10% mixture increase was associated with -5.65% (95% CI: -9.79, -1.28) lower testosterone and -0.09 μIU/mL (95% CI: -0.20, 0.00) lower TSH. Associations with progesterone, testosterone, and FT4 did not differ by fetal sex. However, in women carrying females, we identified an inverted u-shaped relationship of the mixture with estradiol. Additionally, in women carrying females, each 10% increase in the mixture was associated with 1.50% (95% CI: -0.15, 3.18) higher TT4, whereas in women carrying males, the mixture was associated with -1.77% (95% CI: -4.08, 0.58) lower TT4 and -0.18 μIU/mL (95% CI: -0.33, -0.03) lower TSH. We also identified select chemical interactions. CONCLUSION Some biomarkers were associated with early-to-mid pregnancy hormones. There were some sex-specific and non-linear associations. Future studies could consider how these findings relate to pregnancy/birth outcomes.
Collapse
Affiliation(s)
- Brad A Ryva
- Department of Pharmacology and Toxicology, Michigan State University, East Lansing, MI 48824, United States; College of Osteopathic Medicine, Michigan State University, East Lansing, MI 48824, United States; Institute for Integrative Toxicology, Michigan State University, East Lansing, MI 48824, United States
| | - Diana C Pacyga
- Institute for Integrative Toxicology, Michigan State University, East Lansing, MI 48824, United States; Department of Food Science and Human Nutrition, Michigan State University, East Lansing, MI 48824, United States
| | - Kaitlyn Y Anderson
- College of Osteopathic Medicine, Michigan State University, East Lansing, MI 48824, United States
| | - Antonia M Calafat
- Division of Laboratory Sciences, National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, GA 30341, United States
| | - Jason Whalen
- Michigan Diabetes Research Center Chemistry Laboratory, University of Michigan, Ann Arbor, MI 48109, United States
| | - Max T Aung
- Department of Population and Public Health Sciences, Keck School of Medicine, University of Southern California, Los Angeles, CA 90032, United States
| | - Joseph C Gardiner
- Department of Epidemiology and Biostatistics, Michigan State University, East Lansing, MI 48824, United States
| | - Joseph M Braun
- Department of Epidemiology, Brown University, Providence, RI 02912, United States
| | - Susan L Schantz
- The Beckman Institute, University of Illinois, Urbana-Champaign, IL 61801, United States; Department of Comparative Biosciences, University of Illinois, Urbana-Champaign, IL 61802, United States
| | - Rita S Strakovsky
- Institute for Integrative Toxicology, Michigan State University, East Lansing, MI 48824, United States; Department of Food Science and Human Nutrition, Michigan State University, East Lansing, MI 48824, United States.
| |
Collapse
|
3
|
Hu H, Liu X, Zheng Y, He X, Hart J, James P, Laden F, Chen Y, Bian J. Methodological Challenges in Spatial and Contextual Exposome-Health Studies. CRITICAL REVIEWS IN ENVIRONMENTAL SCIENCE AND TECHNOLOGY 2023; 53:827-846. [PMID: 37138645 PMCID: PMC10153069 DOI: 10.1080/10643389.2022.2093595] [Citation(s) in RCA: 10] [Impact Index Per Article: 10.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 05/05/2023]
Abstract
The concept of the exposome encompasses the totality of exposures from a variety of external and internal sources across an individual's life course. The wealth of existing spatial and contextual data makes it appealing to characterize individuals' external exposome to advance our understanding of environmental determinants of health. However, the spatial and contextual exposome is very different from other exposome factors measured at the individual-level as spatial and contextual exposome data are more heterogenous with unique correlation structures and various spatiotemporal scales. These distinctive characteristics lead to multiple unique methodological challenges across different stages of a study. This article provides a review of the existing resources, methods, and tools in the new and developing field for spatial and contextual exposome-health studies focusing on four areas: (1) data engineering, (2) spatiotemporal data linkage, (3) statistical methods for exposome-health association studies, and (4) machine- and deep-learning methods to use spatial and contextual exposome data for disease prediction. A critical analysis of the methodological challenges involved in each of these areas is performed to identify knowledge gaps and address future research needs.
Collapse
Affiliation(s)
- Hui Hu
- Channing Division of Network Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA
| | - Xiaokang Liu
- Department of Biostatistics, Epidemiology, and Informatics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA
| | - Yi Zheng
- Channing Division of Network Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA
| | - Xing He
- Department of Health Outcomes and Biomedical Informatics, College of Medicine, University of Florida, Gainesville, Florida, USA
| | - Jaime Hart
- Channing Division of Network Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA
- Department of Environmental Health, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, USA
| | - Peter James
- Department of Environmental Health, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, USA
- Department of Population Medicine, Harvard Pilgrim Healthcare, Boston, Massachusetts, USA
| | - Francine Laden
- Channing Division of Network Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA
- Department of Environmental Health, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, USA
- Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, USA
| | - Yong Chen
- Department of Biostatistics, Epidemiology, and Informatics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA
| | - Jiang Bian
- Department of Health Outcomes and Biomedical Informatics, College of Medicine, University of Florida, Gainesville, Florida, USA
| |
Collapse
|
4
|
Martenies SE, Hoskovec L, Wilson A, Moore BF, Starling AP, Allshouse WB, Adgate JL, Dabelea D, Magzamen S. Using non-parametric Bayes shrinkage to assess relationships between multiple environmental and social stressors and neonatal size and body composition in the Healthy Start cohort. Environ Health 2022; 21:111. [PMID: 36401268 PMCID: PMC9675112 DOI: 10.1186/s12940-022-00934-z] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/18/2022] [Accepted: 10/30/2022] [Indexed: 06/09/2023]
Abstract
BACKGROUND Both environmental and social factors have been linked to birth weight and adiposity at birth, but few studies consider the effects of exposure mixtures. Our objective was to identify which components of a mixture of neighborhood-level environmental and social exposures were driving associations with birth weight and adiposity at birth in the Healthy Start cohort. METHODS Exposures were assessed at the census tract level and included air pollution, built environment characteristics, and socioeconomic status. Prenatal exposures were assigned based on address at enrollment. Birth weight was measured at delivery and adiposity was measured using air displacement plethysmography within three days. We used non-parametric Bayes shrinkage (NPB) to identify exposures that were associated with our outcomes of interest. NPB models were compared to single-predictor linear regression. We also included generalized additive models (GAM) to assess nonlinear relationships. All regression models were adjusted for individual-level covariates, including maternal age, pre-pregnancy BMI, and smoking. RESULTS Results from NPB models showed most exposures were negatively associated with birth weight, though credible intervals were wide and generally contained zero. However, the NPB model identified an interaction between ozone and temperature on birth weight, and the GAM suggested potential non-linear relationships. For associations between ozone or temperature with birth weight, we observed effect modification by maternal race/ethnicity, where effects were stronger for mothers who identified as a race or ethnicity other than non-Hispanic White. No associations with adiposity at birth were observed. CONCLUSIONS NPB identified prenatal exposures to ozone and temperature as predictors of birth weight, and mothers who identify as a race or ethnicity other than non-Hispanic White might be disproportionately impacted. However, NPB models may have limited applicability when non-linear effects are present. Future work should consider a two-stage approach where NPB is used to reduce dimensionality and alternative approaches examine non-linear effects.
Collapse
Affiliation(s)
- Sheena E Martenies
- Department of Kinesiology and Community Health, University of Illinois at Urbana-Champaign, 906 S Goodwin Ave, M/C 052, Urbana, IL, 61801, USA.
- Division of Nutritional Sciences, University of Illinois at Urbana-Champaign, Urbana, IL, USA.
- Department of Environmental and Radiological Health Sciences, Colorado State University, Fort Collins, CO, USA.
| | - Lauren Hoskovec
- Department of Statistics, Colorado State University, Fort Collins, CO, USA
| | - Ander Wilson
- Department of Statistics, Colorado State University, Fort Collins, CO, USA
| | - Brianna F Moore
- Department of Epidemiology, Colorado School of Public Health, University of Colorado Anschutz Medical Campus, Aurora, CO, USA
| | - Anne P Starling
- Lifecourse Epidemiology of Adiposity and Diabetes (LEAD Center), University of Colorado Anschutz Medical Campus, Aurora, CO, USA
- Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, NC, USA
| | - William B Allshouse
- Department of Environmental and Occupational Health, Colorado School of Public Health, University of Colorado Anschutz Campus, Aurora, CO, USA
| | - John L Adgate
- Department of Environmental and Occupational Health, Colorado School of Public Health, University of Colorado Anschutz Campus, Aurora, CO, USA
| | - Dana Dabelea
- Department of Epidemiology, Colorado School of Public Health, University of Colorado Anschutz Medical Campus, Aurora, CO, USA
- Lifecourse Epidemiology of Adiposity and Diabetes (LEAD Center), University of Colorado Anschutz Medical Campus, Aurora, CO, USA
- Department of Pediatrics, School of Medicine, University of Colorado Anschutz Medical Campus, Aurora, CO, USA
| | - Sheryl Magzamen
- Department of Environmental and Radiological Health Sciences, Colorado State University, Fort Collins, CO, USA
- Department of Epidemiology, Colorado School of Public Health, University of Colorado Anschutz Medical Campus, Aurora, CO, USA
| |
Collapse
|