Kevin Hobbie

Kevin is an expert in environmental health sciences and has more than 15 years of experience specializing in environmental chemistry, epidemiology, and data science. He researches a broad spectrum of topics, including exposure modeling, dose-response modeling, chemical food safety, environmental disaster response, and environmental forensic chemistry.

Since joining our team, Kevin has managed various technical aspects of the systematic review and risk assessment efforts for the United States Environmental Protection Agency (EPA) and the National Institute of Environmental Health Sciences’ (NIEHS) National Toxicology Program. Previously, at Oregon State University, Kevin managed grant and contract awards spanning the United Nations‐Global Environmental Fund, NIEHS, Department of Defense, the state of Oregon, and other private research foundations.

Kevin often engages in projects that require large volumes of data and scrupulous data management. He aids in planning the data collection and analysis while ensuring transparent documentation of systematic review procedures and maintaining high levels of quality control. Backed by a team of technologists and subject matter experts, they deliver results for our clients by utilizing data tools, natural language processing with AI, and data automation to increase the accuracy and efficiency when combing through large bodies of evidence.

In support of the EPA’s Integrated Risk Information System, Kevin led data analysis efforts that supported a rapid exposure-response analysis on hundreds of datasets extracted from epidemiologic studies. This analysis produced over 1,000 benchmark dose models, automated model selection, tabulated model parameters, compiled summary graphics, and produced a large volume of formatted reports using R and R Markdown. Using these results, the EPA prioritized health endpoints for further investigation. Kevin also supported a pre-analysis of selected datasets for a subsequent hierarchical Bayesian meta-regression by determining dose estimates using Monte Carlo simulations while integrating uncertainty with cohort-specific estimates of dose parameters. He established quality control checklists to confirm the accuracy of data processing and consistent application of intake assumptions for dose-response modeling. These analyses underwent review by a National Academies of Sciences scientific review panel and resulted in a series of papers published in 2020.

In support of existing chemicals reviewed under the Frank R. Lautenberg Chemical Safety for the 21st Century Act for the EPA’s Office of Pollution Prevention and Toxics, Kevin supported several data collection and systematic review efforts for problem formulations and risk assessments. A specific challenge in these tasks was collecting and integrating data related to chemical exposures due to the heterogeneous nature of studies that contain useful data and information.

To support exposure assessment of PBT5 (five known Persistent Bioaccumalative and Toxic chemicals) and HBCD (hexabromocyclododecane), Kevin developed data extraction forms in our litstream™ systematic review application to curate exposure data. With a team of programmers they developed a data pipeline to integrate this heterogeneous data, provide exploratory data analysis from various study types with supporting tables and plots, and generate exposure estimates for media used in risk scenarios. This effort involved coordinating with multiple contractors to produce a comprehensive database of available monitoring data that informs exposure estimates to the U.S. general population in various exposure scenarios. This novel approach for integrating monitoring data into a Toxic Substances Control Act exposure assessment underwent initial scientific review by a panel of experts.

Kevin also leads a team of developers that build applications and data tools that support project operations across our service areas. These tools span from data collection, workforce management, data analytics, and document automation. While working with EPA clients, the team enhanced interoperability of the systematic review applications that enabled the multiple offices in the agency to use the right data tools for each task. This increased the speed of delivery through developing application programming interfaces and data pipelines. Kevin acquired additional training in cloud technologies (e.g., Amazon Web Services, Azure, and Google Cloud Platform) to bolster support of our subject matter experts and their roles in the digital transformation of public sector IT with our diverse clients.