EPA Dismantles Research Office, Impacting Chemical Risk Science
In a landmark move driven by industry lobbying, EPA Administrator Lee Zeldin has formally announced the elimination of the agency’s Office of Research and Development (ORD). The decision, unveiled in July 2025, ends decades of independent toxicology, epidemiology, and environmental modeling under the Integrated Risk Information System (IRIS) program and reallocates staff and budget to programmatic offices.
Background: Industry Pressure and Administrative Action
- January 2025: More than 80 industry groups, led by the American Chemistry Council, petition Zeldin to cease ORD’s IRIS assessments.
- March 2025: EPA staff alerted to potential ORD restructuring, triggering internal memos and union objections.
- July 18, 2025: Administrator Zeldin signs an organizational recharting plan estimating $748.8 million in taxpayer savings over five years and a 23% workforce reduction.
Technical Significance of ORD’s IRIS Program
Since its inception in 1985, IRIS has provided quantitative risk values using:
- Physiologically Based Pharmacokinetic (PBPK) Modeling to predict chemical distribution in human tissues.
- In Vitro High-Throughput Screening (HTS) pipelines, including EPA’s ToxCast and Tox21 collaborations.
- Peer-reviewed Dose-Response Assessments applying benchmark dose (BMD) modeling with uncertainty factors for interspecies extrapolation.
ORD’s supercomputing clusters and data lake architecture have supported large-scale in silico toxicology and atmospheric chemistry models, such as the Community Multiscale Air Quality (CMAQ) system.
Latest Developments and Legislative Pushback
In December 2025, the Government Accountability Office (GAO) released an audit warning that the ORD closure could create a data gap in mandated Clean Air Act and Safe Drinking Water Act risk assessments. Concurrently, a bipartisan group in the House introduced the Scientific Integrity Restoration Act to reestablish a centralized research office with statutory protections for scientific autonomy.
Expert Opinions and Industry Reactions
“ORD’s transparent peer‐review workflows set a global standard. Removing that independent firewall risks politicizing core toxicological evaluations,” says Christopher Frey, former EPA ORD assistant administrator and current NC State University associate dean.
“Our members need clarity on chemical safety. ORD was too slow and opaque – we welcome streamlined program office risk tools,” counters an American Chemistry Council spokesperson.
Deeper Analysis
Implications for Chemical Risk Policy and Public Health
Without a unified IRIS framework, individual program offices (air, water, waste) will develop risk values in silos. This fractured approach may lead to inconsistent safety thresholds and impede cross‐media exposure assessments. Public health stakeholders warn of potential delays in updating reference doses for priority pollutants like formaldehyde, ethylene oxide, arsenic, and chloroprene.
Impact on EPA’s Data Infrastructure and HPC Resources
ORD’s in-house high-performance computing (HPC) cluster, comprising 1,200 CPU cores and 200 GPU accelerators, has powered advanced climate and exposure models. Migrating those workloads to smaller program office servers or third-party cloud instances raises concerns over data security, GxP compliance, and long-term operational costs. Cloud migration studies estimate a 15–20% premium in licensing and egress fees.
Alternatives and Future Outlook
- Academia‐led Consortia: University research centers are forming coalitions to maintain open‐source IRIS‐style databases.
- Public–Private Partnerships: Proposals surfaced in early 2026 to leverage NOAA and NIH computing assets for cross-agency toxicology modeling.
- Legislative Remedies: The proposed Scientific Integrity Restoration Act could reinstigate a protected research division within EPA.
Conclusion
The shuttering of ORD represents a seismic shift in how chemical risk science informs regulatory policy. As program offices absorb research functions, experts caution that the loss of a centralized, peer‐reviewed system could undermine the agency’s ability to deliver consistent, transparent, and scientifically robust assessments—potentially delaying public health protections and sowing uncertainty in environmental markets.