Pre-Meeting Workshops
Full Day Workshops (08:00 - 15:30)
G. Konstantinoudis1, R. Parks2
1 Imperial College London, Grantham Institute for Climate Change and the Environment, London, United Kingdom
2 Columbia University, Department of Environmental Health Sciences, New York, New York, United States of America
In environmental epidemiology, spatial and spatiotemporal methods are increasingly used with high-resolution environmental and health data, which allows detailed insights into associations between exposures and outcomes. In this context, Bayesian methods provide a natural setting to incorporate uncertainty and prior knowledge, borrow information between neighbouring units, and create hierarchical structures. However, appropriate usage of these methods requires careful consideration and knowledge of building technical models, which can be intimidating to the uninitiated user.
During this full-day course, our aim is to introduce the ideas of Bayesian inference and modelling for spatiotemporal environmental data. The workshop is designed to be as approachable and friendly as possible while still providing technical and practical know-how.
The objectives of the workshop are:
- An Introduction to Bayesian inference
- An Introduction to hierarchical model structures
- An Introduction to spatiotemporal priors
- Applications in real world environmental data and health outcomes
We will focus on three illustrative case studies.
In the first case study, we will evaluate the effect of long-term air-pollution exposure on COVID-19 mortality in England and examine the crucial role of spatial autocorrelation.
In the second case study, we will estimate the non-linear effect of temperature in all-cause mortality in Italy and examine how this effect has changed over time and varies in space using spatiotemporal Gaussian processes.
In the third case study, we will show how spatiotemporal models can be used to evaluate excess deaths due to events such as heatwaves in Switzerland and ways to propagate the different sources of uncertainty.
All analysis will be performed in R with the NIMBLE software and data and code will be made available for future studies.
By the end of this workshop, attendants shall develop an understanding of Bayesian spatiotemporal models and be able to write their first spatiotemporal model in NIMBLE.
D. I. Walker1, G. W. Miller2, A. S. Young1, D. Liang1, X. Hu1
1 Emory University, Gangarosa Department of Environmental Health, Atlanta, Georgia, United States of America
2 Columbia University, Department of Environmental Health, New York, New York, United States of America
Untargeted high-resolution mass spectrometry (HRMS) has become a key platform for assessing the exposome, and its use is rapidly growing in environmental epidemiology and biomedical research for measurement of thousands of chemicals in human populations. While these methods provide a powerful strategy for comprehensively studying the role of environmental factors in human disease, the resulting data is complex, and proper interpretation and analysis requires an understanding of underlying instrumental methods and data pre-processing approaches. The objective of this full-day workshop is to provide participants with training in all aspects of the untargeted HRMS workflow for characterizing the exposome and will be taught by internationally recognized experts in the development and application of untargeted HRMS approaches for studying the exposome. Theoretical sessions will include the role of untargeted HRMS for exposome research and a detailed overview of the untargeted HRMS workflow, including instrumentation platforms, open-source methods for data pre-processing, and strategies for compound identification/annotation and reporting. Practical sessions will focus on the latest advances in data analysis methods for untargeted HRMS, including hands-on experience with single-variable and mixture-based methods, combining exposome-wide association studies with high-dimensional mediation analysis in a meet-in-the-middle framework for studying the intersection of exposures, biological response, and disease, and multi-omic approaches for integrating the exposome within biological networks. Upon completion of the workshop, participants will have developed expertise in the theoretical principles underlying the use of advanced analytical chemistry methods for measuring the exposome and gain hands-on experience in the application of cutting-edge data analysis methods that maximize information gained from untargeted HRMS measurements, empowering attendees to apply these methods to their own research projects.
J. Wambaugh1, C. Ring1, P. Kruse2, K. Phillips1, K. Isaacs1
1 U.S. Environmental Protection Agency, Office of Research and Development / Center for Computational Toxicology and Exposure, Research Triangle Park, North Carolina, United States of America
2 Oak Ridge Institute for Science and Education, Oak Ridge, Tennessee, United States of America
The EPA’s Office of Research and Development (ORD) has developed a suite of tools to aid in screening exposure and hazard of chemicals relative to the general population. These tools are accessible graphical user interfaces (GUI) and application programming interfaces (APIs). This workshop will introduce users to the types of data available in these tools and then familiarize them with both the GUI and API versions of the tools. In addition, this course aims to give attendees who may have experience in risk assessment or exposure science, but who may lack formal training in computer programming, the time and ability to learn the requisite skills needed to utilize the APIs, which are becoming an increasingly popular way to deliver and access data. After a brief introduction into the type of information available in these tools, this workshop will cover web-based GUIs developed by EPA: namely, the CompTox Chemicals Dashboard (CCD) and the Chemical Exposure Knowledgebase (ChemExpo). After this, there will be hands-on time to teach attendees how to download and install a scripting programming language (attendees may choose either R or Python). This will be followed by a basic introduction on how to write elementary code (e.g., reading, writing, filtering spreadsheet files) and examples of how to access EPA’s Computational Toxicology and Exposure (CTX) and Hazard Comparison Dashboard (HCD) APIs using available open-source tools. Finally, the course will give attendees the opportunity to perform two screening-level case studies: a single chemical assessment and a prioritization activity. These assessments will be carried out using CCD and ChemExpo and then will be performed using the APIs. Using both methods will 1) reinforce how to use EPA’s data and tools and 2) allow comparison of the two types of interfaces to help attendees determine which interface may be more suitable to their needs.
Half Day Workshops (Morning, 08:00 - 11:30)
K. Christensen1, R. Nachman1, P. Bommarito1, L. Vegosen1, F. Branch1, M. Powers1
1 United States Environmental Protection Agency, Washington, DC, United States of America
The identification and evaluation of information used in health assessments increasingly relies upon systematic review methods to ensure transparency, consistency, and reproducibility. Systematic reviews typically include several steps, such as scoping and problem formulation, literature search and screening, and study evaluation and data extraction. Components of study evaluation include risk of bias (RoB) and sensitivity. RoB assesses the internal validity of the study, whereas sensitivity is the ability of the study to detect a true association where one exists. There are multiple frameworks that may be used to structure a RoB evaluation, which have been developed to increase transparency and reduce subjectivity in reviews. Selection of a specific tool depends on the goal of the systematic review and available resources. This workshop will provide an overview of identifying the relevant body of evidence for a systematic review; evaluating RoB and sensitivity for observational epidemiology studies; briefly describe commonly encountered frameworks such as ROBINS-E, Navigation Guide, OHAT, the U.S. Environmental Protection Agency’s (US EPA) Toxic Substances Control Act (TSCA) and Integrated Risk Information System (IRIS) frameworks; and discuss tailoring a framework to fit the needs of the systematic review. The workshop will also include a hands-on exercise tailoring and applying a RoB framework for a specific research or assessment need using the US EPA IRIS framework. This workshop is designed for epidemiologists and other researchers working with human data, to learn how their research may be evaluated during the systematic review process, as well as those directly engaged in conducting systematic reviews in the research or practice arena and/or using epidemiologic research in risk assessment.
*Disclaimer: The views expressed in this abstract are those of the authors and do not necessarily reflect the views or policies of the U.S. EPA.
B. Welch1, A. Keil2, M. Spaur2
1 University of Nevada, Reno, School of Public Health, Reno, Nevada, United States of America
2 National Cancer Institute, Occupational and Environmental Epidemiology Branch, Rockville, Maryland, United States of America
Environmental epidemiologists are often interested in the health effects of chemicals that are, or may become, subject to policy and regulations aimed at reducing population-level exposures. Increasingly, the field of environmental epidemiology needs tools that can directly inform the public health impacts of future interventions. Further, there is an ongoing need for epidemiologic tools to address the complex reality of exposure to multiple contaminants in a way that informs decisions in an interpretable way. Causal inference methods, such as g-computation, are analytic approaches that can address both of these needs. In this workshop, we will provide an overview and practical advice on how to use g-computation, a methodological approach that leverages standard regression tools, to assess exposure-health relationships in terms of hypothetical interventions that can mimic potential public health actions. Hypothetical interventions can be crafted to address questions that may provide more insight than typical regression output in epidemiologic studies, such as:
- How would disease risk change if we could reduce all exposures within a mixture?
- Could health disparities be reduced or eliminated by ensuring that all groups receive similar exposures?
Using a combination of theory and worked examples, we will provide attendees the foundational tools to address such questions. Workshop material will cover theoretical topics such as: a basic introduction to the potential outcome framework of causal inference, and causal inference for exposure mixtures. Attendees will be given a synthetic dataset and sample R code for analysis. We will also consider practical issues such as: data management in causal analysis, sensitivity analyses to consider, and basic use of non-parametric bootstrapping to calculate confidence intervals. Deliverables for attendees include: a) an understanding of how to frame a hypothetical intervention; and b) experience using code to analyze a hypothetical intervention.
A. Miller1, M. Conway2, K. Messier3, T. Castranio4, A. Liu1, 6, K. Lane5, E. Gause5
1 National Institute of Environmental Health Sciences, Division of Extramural Research and Training, Bethesda, Maryland, United States of America
2 National Institute of Environmental Health Sciences, Office of Data Science, Durham, North Carolina, United States of America
3 National Institute of Environmental Health Sciences, Division of Translational Toxicology, Durham, North Carolina, United States of America
4 National Institute of Environmental Health Sciences, Division of Extramural Research and Training, Durham, North Carolina, United States of America
5 Boston University School of Public Health, Environmental Health, Boston, Massachusetts, United States of America
6 Kelly Government Solutions, Rockville, Maryland, United States of America
Researchers investigating the health impacts of environmental exposures require access to data from multiple disciplines, ranging from earth and climate sciences to social determinants of health to clinical health. Though a wealth of data from numerous organizations exists, they often are generated for purposes other than scientific research and therefore reside in disparate locations, lack standardized ontologies and approaches for harmonization; require linkage; and are subject to inconsistent rules for access and use. This half-day workshop aims to introduce resources and tools developed by the NIH Climate and Health Outcomes Research Data Systems (CHORDS) project and the CAFÉ Research Coordinating Center to facilitate such health research using geospatial and environmental exposure data.
The outline of the workshop is as follows:
- Overview of the CHORDS data ecosystem and showcase of the curated data catalog and relevant resources, such as the climate change and human health literature portal and glossary
- Overview of the CAFÉ data management function, Harvard Dataverse repository, and research tools and educational resources
- Tutorial on the ‘amadeus’ software tool using a case example
- Tutorial of the processing and analysis guides available from CAFÉ
- Question and answer session
- Focus group discussion of climate and environmental data standards efforts
Following the completion of the workshop, attendants will have a better understanding of the diverse geospatial and environmental exposure data, tools, and resources; evolving approaches for linking and harmonizing data; and considerations for data standards in the context of epidemiologic and clinical research.
Half Day Workshops (Afternoon, 12:00 - 15:30)
R. M. Shaffer1, E. Copeland2, A. Kostant2, P. Pant3
1 U.S. Environmental Protection Agency, Washington DC, United States of America
2 Science Communication Network, Bethesda, Maryland, United States of America
3 Health Effects Institute, Boston, Massachusetts, United States of America
Effective communication of science concepts and scientific research is an essential skill for environmental epidemiologists and exposure scientists. Science communication makes our science more accessible, collaborative, and impactful. This workshop will provide theoretical and practical guidance to improve science communication across multiple platforms and across all career stages.
The first part of the workshop will include two trainings and one interactive breakout session. First, Science Communication Network (SCN) experts will provide “Media Training 101,” which will teach the nuances of communicating environmental health science to the media and other public audiences. An essential component of this training will be a focus on delivery of key research findings in an approachable manner while protecting scientific credibility. Following this presentation, participants will practice incorporating aspects of this training using ISEE/ISES-related examples in small breakout groups. Next, SCN experts will provide “Social Media for Scientists 101” training, which will include a discussion of best practices for scientists using social media in order to maximize their comfort and credibility using a variety of platforms.
In the second part of the workshop, two ISEE and ISES members will provide an overview of best practices based on real-world experiences in environmental health science communication, including in low and middle-income countries. The session will close with a discussion of opportunities and resources to advance skills in science communication and a Q&A with participants.
These trainings, coupled with the planned interactive break-out and discussion sessions, will provide essential guidance and practice for a world of increasing polarization and rapidly changing media and social media landscapes.
*Disclaimer: The views expressed in this abstract are those of the authors and do not necessarily reflect the views or policies of the U.S. EPA.
T. Benmarhnia1, Y. Ma1
1 University of California San Diego, Scripps Institution of Oceanography, La Jolla, California, United States of America
Quasi-experimental methods have been increasingly applied in environmental epidemiology to evaluate the effectiveness of health interventions or to assess the health impacts of extreme weather events. Capitalizing on the timing of natural experiments, difference-in-differences and synthetic control methods can be used to estimate causal effects under specific identification assumptions, especially when it is impractical or infeasible to conduct a randomized controlled trial. Such methods require one or more control groups not exposed to the event; however, suitable control groups are not always available. In this context, the interrupted time series (ITS) design offers a valuable alternative.
The ITS design is well suited to evaluating the population-level health effects of events or interventions that occur at a clearly defined time point in a single exposed or intervened population. This design extrapolates the time-series data from the pre-event phase to generate a counterfactual trend that represents what would have been observed (under specific assumptions) in the absence of the event. This method has been widely used in public health settings, including the evaluation of health impacts of environmental policies or extreme weather events such as floods or wildfires.
In this workshop, we will first provide a theoretical overview of both traditional and recently proposed ITS modeling approaches under the potential outcomes framework, spanning from standard one- and two-stage models to machine learning-integrated methods and extensions for staggered intervention scenarios. We will discuss the assumptions, model specifications, advantages, and potential methodological challenges of these approaches. We will focus on examples related to air quality warning systems and wildfire events. In the second part of this workshop, we will showcase multiple ITS modeling approaches using R programming language, with examples from environmental epidemiology to facilitate translation to real-world data analysis. We will guide attendees to work on hands-on programming exercises.
K. Pollitt1, E. Lin1
1 Yale School of Public Health, Environmental Health Sciences, New Haven, Connecticut, United States of America
Capturing the diverse and dynamic environmental influences that we are exposed to is challenging. Wristbands have recently emerged as wearable tools that can provide personalized measures of chemical and biological exposures and track how these exposures change over time. The lightweight and lower cost design of wristbands have enabled exposures to be assessed in remote communities, vulnerable populations, and large-scale cohorts. This workshop will provide a comprehensive guide to using wristbands for exposure assessment, from the fundamentals of passive sampling to practical applications of the tool and an opportunity to work hands-on with real data. Participants will gain insight on study design and logistics, including available wristband technologies, best practices for deploying these wearable tools, optimal wear durations, and essential protocols for quality control and assurance. The session will also discuss mass spectrometry methods for targeted and non-targeted chemical analysis and participants will gain experience with data handling, analysis, and integration with various EPA databases, and interpretation. We welcome investigators at all career stages to attend. A basic understanding of R is recommended to maximize engagement in the workshop demonstrations.