Spotlight On: Health & Environment Systematic Review Capabilities
Abt Global has helped the Environmental Protection Agency (EPA), the Agency for Toxic Substances and Disease Registry (ATSDR), and other clients develop methods for—and implement—systematic reviews. We are well-versed in using the tools and methods required to conduct a transparent and defensible product.
Our Systematic Review Capabilities
Abt’s deep understanding of the systematic literature review process is based on our hands-on experience screening studies, extracting data, conducting bias assessments, and summarizing results for multiple clients. We understand there are standardized methods, tools, and approaches that are utilized in order to ensure transparency in not only the literature review process but also in hazard characterization.
Our Health and Environment team has developed many different literature review products, ranging from full systematic reviews and meta-analyses to quick-turnaround summaries, reference lists, and annotated bibliographies. We work with clients to identify and tailor the most appropriate format for each work effort.
Approach
To conduct successful literature and document reviews, our team draws on Abt’s internal capabilities, which include digital subscriptions to a wide range of information sources and peer-reviewed journals; capable staff who can systematically sort through hundreds or thousands of potential references explicitly documenting the inclusion or exclusion of the potential references by a predetermined criteria; subject matter experts who can summarize literature coherently; and experienced technical writers, editors, graphic specialists, and web designers who can effectively communicate findings through print or online reports. Additionally, we use both licensed and Abt-created tools and templates to facilitate and manage every step of the process. Our data scientists have developed Artificial Intelligence and Natural Language Processing solutions to drive efficiencies. We tailor our literature searches and document search methods to the needs of the project and adhere to a structured approach to provide an unbiased and high-end product. We understand the need for a predetermined methodology, appropriate documentation, the integration of evidence, and unbiased presentation of findings.
Expertise
Abt currently holds more than $15 million in contracts related to systematic review. Further, we are familiar with EPA’s approach to conducting systematic reviews under the amended Toxic Substance Control Act and potential updates to this process. We also appreciate the immense amount of work required to complete these reviews and continuously strive to increase efficiency.
Abt is experienced in using tools for systematic reviews, including:
- DistillerSR,
- Health Assessment Workspace Collaborative (HAWC),
- Swift-Review and Swift-Active Screener,
- Abstrackr,
- Tableau,
- JIRA, and
- Several coding languages for data processing (i.e., R, python).
Through our work, we have familiarity and have used defined methods while conducting reviews, such as:
- the Navigation Guide
- the National Toxicology Program’s Office of Health Assessment and Translation’s (OHAT) handbook,
- Draft Systematic Review Protocol Supporting TSCA Risk Evaluations for Chemical Substances (EPA), and
- the National Academy of Science recommendations based on reviews of EPA methods and guidance documents.
Selected Experience
Development of Data Quality Criteria to Evaluate Exposure Data in Environmental Epidemiology Studies
Client: U.S. EPA Office of Pollution Prevention and Toxics (Under Subcontract to Versar, Inc.)
To support the development of methods for conducting systematic literature reviews for high-priority chemicals under the amended Toxic Substance Control Act, Abt Global (under subcontract to Versar Inc.) developed data quality criteria to evaluate the exposure data presented in environmental epidemiological studies. To develop these criteria, an evaluation of EPA’s General Assessment Factors (the biomonitoring, environmental epidemiology, and short-lived chemicals instrument), and the OHAT’s risk of bias (RoB) tool were necessary. Abt utilized these tools to inform the development of evaluation metrics by which to assess the exposure assessment methods used in environmental epidemiological studies. Once the criteria were agreed upon with EPA, we tested them with several studies and provided input to EPA on the lessons learned from conducting the test evaluations.
Support for the Development of Toxicological Profiles
Client: Agency for Toxic Substances and Disease Registry
Abt is supporting ATSDR in all facets of developing toxicological profiles (Tox Profiles) for contaminants found at national priority list sites, including for some priority list chemicals such as 1,2 dichloroethane. This work includes defining search strings, searching various databases for toxicity information, and synthesizing the literature in a comprehensive report. We have implemented systematic reviews for several of the chemicals, for which we have then developed Tox Profiles. To implement the reviews, we begin by uploading our search results into DistillerSR and use our custom-made forms to conduct title, abstract, and full-text screening. Data extraction is conducted for studies that are included after full-text review, using Abt-developed DistillerSR forms. For select endpoints, risk of bias and confidence assessments are conducted using the RoB tool and ATSDR’s guidelines on confidence assessment. The data evaluations are then integrated into conclusions on chemical hazard. Through our partnership with ATSDR, we have screened tens of thousands of articles, defined inclusion and exclusion criteria, and created standardized processes and forms that can be tuned to other projects that implement systematic review for evaluating chemical hazard. Additionally, we have developed automated quality assurance procedures for both our data and previously collected Tox Profile data, introduced automations in the process, and developed detailed data visualizations of the human health evidence in Tableau.
Evaluation of the Data Gaps in the Literature on PFAS and Immune Effects
Client: Federal Client
Abt conducted a literature review of the immunosuppressive effects of per- and polyfluoroalkyl substances (PFAS) and assessed National Health and Nutrition Examination Survey (NHANES) data to identify whether available data may address key gaps in the literature. Working with Abt’s in-house librarian, search strings were developed to identify literature that evaluated 15 PFAS chemicals and potential impacts on the immune system. Using HAWC, Abt conducted title, abstract, and full-text screening of approximately 2,000 toxicological and epidemiological articles. For articles that were included after full-text review, data was extracted in Abt-developed spreadsheets. The Abt team then synthesized the studies on potential immunologic impacts associated with PFAS exposure in the form of evidence tables and three separate reports: a summary of toxicological evidence, a summary of epidemiological evidence, and a synthesis of data to outline data gaps.
Key Staff
Meghan Lynch
Health & Environment
Dr. Lynch has over 20 years of experience and specializes in human health risk assessment. She has served as project manager on systematic reviews involving various environmental contaminants. Lynch is particularly particularly interested in benefits estimation which requires synthesizing and translating the findings from toxicological and epidemiological literature to estimates that are useful for benefits analysis. This involves adapting and running probabilistic models to estimate risks and using EPA’s Benchmark Dose Modeling Software. Lynch has a MPH and D.Sc. in environmental health from Boston University School of Public Health. Visit Meghan's full bio.
Chantel Nicolas
Dr. Chantel Nicolas is a chemist with over 15 years of multidisciplinary experience, and she specializes in data science, computational toxicology, chemical risk assessments, and regulatory policy. Dr. Nicolas designs reproducible quantitative analyses of complex data related to chemical risk assessments and systematic reviews. As an exposure scientist with EPA, Dr. Nicolas has developed and led exposure systematic review efforts for TSCA “Next 20” work plan chemicals, which included a protocol for using AI tools to prioritize relevant literature and reduce the level of effort in the screening process. Dr. Nicolas has led a published TSCA exposure assessment (HBCD), developed a published risk assessment prioritization dashboard (TTC Data Mart), and led various published articles relevant for structure-activity relationships for estimating chemical hazards and exposures. She is skilled in DistillerSR, SWIFT-Review, SWIFT-Active Screener, HAWC, Tableau, JIRA, R, and Python.
Sandra Jo Wilson, Ph.D.
Social and Economic Policy
Dr. Wilson is an internationally known expert in research synthesis and meta-analysis. This includes designing, conducting, and disseminating rigorous research studies and systematic reviews. From 2010 to 2017, she was an editor at Campbell Systematic Reviews, where she provided assistance to grantees on systematic review design and facilitated trainings on this topic. Dr. Wilson was awarded the 2020 Society for Prevention Research’s Nan Tobler Award for Review of the Prevention Science Literature. Wilson has a Ph.D. in policy development and program evaluation from Vanderbilt University. Visit Sandra's full bio.
Jenna Harder, Ph.D.
Dr. Jenna Harder is a data scientist with expertise in automation, AI, and statistical analysis. She is experienced at applying advanced methods such as natural language processing, computational modeling, webscraping, and machine learning to creatively solve problems and streamline workflows. In addition to leading data science work, she has experience with activities across the research lifecycle including conducting systematic literature reviews, designing survey and experimental studies, and writing and peer-reviewing scientific articles. She received her PhD in Psychology with a concentration in Quantitative Methods in 2020 from Michigan State University, where her work included developing computational models of decision-making. Dr. Harder joined Abt in July of 2020.