OIG Reports that EPA “Lacks Complete Guidance for the New Chemicals Program to Ensure Consistency and Transparency in Decisions”

On August 2, 2023, the U.S. Environmental Protection Agency’s (EPA) Office of Inspector General (OIG) released a report entitled The EPA Lacks Complete Guidance for the New Chemicals Program to Ensure Consistency and Transparency in Decisions. According to OIG, EPA’s New Chemicals Division (NCD) “lacks assurance that the new chemicals review process operates as intended and achieves its objective to protect human health and the environment.” OIG made four recommendations, including that EPA develop, update, and issue final guidance for the New Chemicals Program; assess and update NCD’s recordkeeping applications, as needed; and address workload issues. EPA agreed to all four recommendations, “which are resolved with corrective actions pending.”

OIG initiated the audit to determine the extent to which EPA is using and complying with applicable records management requirements, quality assurance requirements, and employee performance standards to review and approve new chemicals under the Toxic Substances Control Act (TSCA) to manage human health and environmental risks. OIG states that it conducted the audit in response to complaints submitted to the OIG hotline in summer 2021 regarding EPA’s new chemicals review process. According to OIG, the complaints “expressed concerns about recordkeeping and quality assurance activities, including potential violations of the EPA’s Records Management Policy, improper quality assurance processes, and the prioritization of reviews of new chemicals over the development of standard operating procedures.” The hotline complaints “also expressed concerns related to employee performance standards, such as the use of the TSCA statutory deadlines to perform reviews quickly rather than for the purpose of protecting human health and the environment.” OIG used these three main areas of concern — recordkeeping, quality assurance, and employee performance standards — to develop its audit objective.

According to OIG, EPA did not comply with applicable quality assurance and recordkeeping requirements for the New Chemicals Program. Specifically, NCD did not have, update, or issue final guidance for many of the activities that comprise the new chemicals review process. OIG states that according to EPA’s Guidance for Preparing Standard Operating Procedures, “the development and use of standard operating procedures are integral parts of a successful quality system, as they provide individuals with the information to perform a job properly, and they facilitate consistency in the quality and integrity of products or end results.”

OIG’s report includes the following table of guidance documents not developed, updated, or issued in final as of September 2022.

Risk Assessment Area Covered	Guidance Document Name	Deficiencies Identified
Human health and risk assessment	Evaluation Protocol — Exposure Quality of Inhalation Toxicity Studies	Outdated; not final.
Human health and risk assessment	Checklist for QCing Human Health Risk Assessments (Part Bs)	Outdated; not final.
Exposure assessment	New Chemical Workflow Summary RAD Exposure Assessor	Outdated; not final.
Exposure assessment	CEM Use Frequency and Inhalation Concentration	No file date; unable to determine if current; not final.
Environmental hazard assessment	Hazard and Risk Determination Language for Ecotox Report in NCR	Outdated.
Environmental hazard assessment	Options when Determining a Toxicity (T) Score a	Not final.
Exposure assessment	New Chemicals Step Action Guide	Guidance document did not exist prior to 2020. Staff developed guidance in 2020, but the document is not signed and final.
Scoping meetings	—	Guidance document did not exist. According to management, the guidance was in development.

Notes: CEM = Consumer Exposure Model; NCR = New Chemical Review; RAD = Risk Assessment Division.

Furthermore, OIG states that NCD had issued final guidance for recordkeeping of scoping meetings, and it used multiple recordkeeping applications that were not integrated and were frequently inaccessible. EPA’s Records Management Policy requires that each EPA program office create, receive, and maintain records providing adequate and proper documentation and evidence of EPA’s activities and decisions. OIG notes that in addition, prior to September 2021, NCD’s TSCA recordkeeping applications did not track edits to documents that were developed during the new chemicals review process to support EPA’s decisions about the risks of new chemicals.

OIG states that these deficiencies existed because NCD lacked sufficient staff resources to both conduct reviews within statutory timeframes and develop and issue final guidance for conducting the activities that comprise the review process. Although EPA has the authority to collect fees to offset the costs of implementing the requirements under TSCA, “it has fallen short of collecting the amount of fees it originally projected.” According to OIG, the absence of final guidance and the lack of resources “increase the risk that the new chemicals review process does not meet its legislative intent to prevent unreasonable risk to human health and the environment.”

OIG states that according to NCD’s March 2022 workload and workforce analysis, “NCD lacks the necessary staff resources to update and finalize the New Chemicals Program guidance and improve upon its TSCA information technology systems.” According to OIG, the Office of Chemical Safety and Pollution Prevention (OCSPP) “can use or redistribute additional staff and financial resources awarded to TSCA programs and make staff adjustments to balance the New Chemicals Program’s workload to better operate as intended — in other words, to conduct new chemical reviews that ensure the protection of human health and the environment, while also meeting statutory deadlines.”

According to OIG, hotline complainants alleged that the TSCA statutory deadline was used to affect adversely performance reviews of staff, that staff were evaluated based on the percentage of new chemicals reviews completed within the statutory deadline period, and that staff are rewarded for quickly finishing cases that do not find risks instead of for protecting human health and the environment. OIG states that to address these complaints, it assessed whether NCD’s employee performance standards included a time-bound requirement to complete new chemicals reviews. OIG “found no evidence that the EPA uses employee performance standards to explicitly measure whether employees satisfy the TSCA statutory 90-day review requirement.”

OIG recommends that the Assistant Administrator for OCSPP:

1. Develop and implement a plan to review regularly NCD’s guidance documents, including standard operating procedures, to ensure that all required guidance is developed, current, signed, and issued in final;
    
2. Develop a process to assess periodically the effectiveness of NCD’s official recordkeeping system within the TSCA confidential business information (CBI) systems and update the applications and systems as needed, while maintaining the use of version controls to preserve edits made to records;
    
3. Develop and implement a plan to identify root causes for frequent technical issues and prioritize the creation and implementation of plans of action and milestones based on the severity of the technical issues within the TSCA CBI systems; and
    
4. Conduct periodic reviews of NCD’s workforce and workload analysis, and update as needed, to regularly balance NCD’s workload with the staff resources needed to execute new chemicals review work, including updating and issuing final guidance and maintaining and updating TSCA CBI systems.

According to OIG, OCSPP agreed with its recommendations and proposed corrective actions and estimated completion dates that OIG believes will satisfy the intent of the recommendations. Therefore, OIG considers all recommendations resolved, with corrective action pending. OCSPP agreed with Recommendation 1 to create a plan to review regularly NCD’s guidance documents and stated that the plan will be based on progress and approaches developed to date.

In its response to Recommendation 2, OCSPP stated that the Office of Pollution Prevention and Toxics (OPPT) and the Office of Program Support will regularly meet to identify improvements needed for future versions of the New Chemical Review application. According to OIG, the Office of Program Support will also report annually to the Assistant Administrator for OCSPP on the effectiveness of the New Chemical Review application functions to ensure that recordkeeping within the TSCA CBI systems is consistent with its recommendation.

In response to Recommendation 3, OCSPP stated that the Office of Program Support will develop and initiate a plan as stipulated in the recommendation and that the plan will include:

• A risk assessment to identify root causes for frequent technical issues to the TSCA CBI LAN availability;
   
• A mitigation plan to prioritize hardware and software configuration changes to increase availability of the system and applications; and
   
• A monitoring plan to control the improvements, detect disruptions early and often, and maintain stability within the system.

For Recommendation 4, OCSPP provided two corrective actions: that the Office of Program Support, Mission Support Division, and NCD will develop a schedule for periodic workforce and workload analysis by December 31, 2023, and that the Office of Program Support will update and issue final guidance and a schedule for maintaining and updating the TSCA CBI systems by December 31, 2023.

Commentary

Most importantly, contrary to the allegations that lead to the investigation, OIG did not “find evidence that employees are rewarded for quickly completing risk assessments that do not find risks instead of with the goal of protecting human health and the environment.” This continues the findings, including the National Archives and Records Administration (NARA) investigation related to recordkeeping, that the allegations of misconduct, including by current and former senior career officials, are not supported. We appreciate that EPA staff work hard and with integrity, and this OIG report disabuses any view to the contrary.

We agree that NCD lacks sufficient quality assurance and quality controls (QA/QC) in its new chemicals assessments. Our experience is that it is not unusual for the outcome of a new chemical submission to depend on which assessor reviews it. We have no doubt that the staff turnover, brain drain, and short timelines contribute to the lack of consistency and present significant challenges. These challenges, however, make robust standard operating procedures and QA/QC essential. As with nearly all aspects of the Frank R. Lautenberg Chemical Safety for the 21st Century Act (Lautenberg) implementation, the lack of time for EPA to develop new policies and procedures and then pivot to their comprehensive and seamless implementation were aspirational goals that EPA had no chance to achieve because of the design of the law, not for want of effort. Seven years in, however, these measures should be baked into the operating practices of NCD.

We also agree that EPA must meet recordkeeping requirements. OIG found a lack of a central system for such recordkeeping, not that records of versions were not maintained as alleged.

An interesting aspect of the report is its determination that OIG found evidence that “the standards explicitly address the 90-day requirement of new chemical reviews.” As readers know, the long delays with new chemicals review have been a particularly contentious aspect of Lautenberg implementation.

While it is true that some reviews can be completed more quickly than others, it is also true that the statute is explicit that Congress expects EPA to complete those reviews within 90 days, or at least no more than 180 days. Lautenberg left this timeframe undisturbed from its 1976 origin. Reasonable people may disagree as to whether Congress’s expectation was rooted in reality or a vestigial oversight given the complexity of new chemicals reviews. Few would disagree, however, that the reviews continue to take far too long and contribute to significant commercial disruption. Reviews must be targeted for timely completion. Perhaps EPA could consider time-limited performance standards; that is, a metric of performance could include the review time for cases completed. It is reasonable to expect that an employee complete assigned cases in an appropriate amount of time, weighing the complexity of the case and the amount of information required to be reviewed in the assessment. This is a separate consideration of workload (how many cases are assigned to an individual). Such performance standards are critical to understanding the workforce needs and will necessarily inform TSCA fees.

It is also important to expect that an employee’s review of assigned cases meet the OPPT standards. Employees that do not follow OPPT standards are objectively not performing satisfactorily, and such conduct should be corrected rather than ignored.

We continue to be baffled by EPA’s apparent characterization that, prior to Lautenberg, OPPT only reviewed 20 percent of notices. This mischaracterizes the new chemicals review process prior to 2016. That a chemical was “dropped” from further review does not mean it was not reviewed. We suspect (although EPA has never confirmed) that the “20 percent” metric was the number of “standard review” cases. Standard review cases were de novo reviews performed when EPA lacked sufficient experience with the substance, or class of substances, to complete a categorical review. The other cases were category cases based on EPA’s new chemicals categories — groups of substances with common functional groups (such as acrylates and isocyanates) that EPA had extensive experience reviewing and could efficiently review new chemicals against those categories and take any necessary actions (such as issuing a “category 5(e) order” or a non-order significant new use rule (SNUR)) based on that experience.

The Biden Administration has appropriately embraced the utility of a category approach and has been using it to review closely related premanufacture notifications (PMN) more efficiently through the new chemicals process. Drop cases were (roughly) the equivalent of “not likely” determinations under Lautenberg. EPA came to its conclusion about the hazard and lack of potential risk and concluded that regulation was not necessary. Those cases were dropped from further review and action and the manufacturer had to wait out the 90-day review period.

Under Lautenberg, EPA must make its determination and record and disseminate that determination, either in a written “not likely” determination document or in a regulation (presumably an order). The lack of a written, public determination for each case prior to Lautenberg is not evidence that the case was not reviewed. EPA produced tens of thousands of new chemicals reports, including “Focus decisions,” on dropped cases. Reasonable people can debate whether that process was sufficiently transparent and whether EPA was making appropriate risk decisions, but to suggest that the cases were not reviewed is false.

There are several other errors or misstatements in the report. These are relatively minor in terms of the OIG report’s conclusions, but accuracy in an OIG report matters. For example, contrary to the report’s assertion, EPA did have time-bound requirements prior to Lautenberg. EPA was required to act within 90 days to propose limits on a PMN (or significant new use notice (SNUN)) or the submitter could proceed to commence commercial manufacture. EPA was not required to review each case, but in our experience, it did.

The report’s statement that EPA received 4,514 PMNs is also incorrect. We presume that number represents all types of Section 5 submissions, PMNs, SNUNs, microbial commercial activity notices (MCAN), low volume exemptions (LVE), low release and exposure exemptions (LoREX), test marketing exemptions (TME), TSCA environmental release applications (TERA), and modifications of exemption notices.

Similarly, the report misstates the NCD review process. The report states that scoping “takes six days.” We presume that this is OIG’s interpretation of the Figure 1 diagram in which Scoping Meeting is labeled “Days 17-22.” Based on our experience, the “Days” number is not the number of days the activity requires but the number of days from “Day 1” that the activity is expected to occur. This is evident in the overlap of the “days” boxes for activities. The Hazard Meeting is targeted for days “14-19.” The Scoping Meeting is targeted for “Days 17-22,” but the Scoping Meeting cannot occur until the Hazard Meeting is complete, so if “Hazard Meeting” is intended to take six days (from Day 14 to Day 19), the earliest Scoping could occur is Day 20. This is a minor error, but it suggests that OIG has not fully understood the New Chemicals Program.

We agree with the report’s major conclusions. We hope stakeholders agree that standard procedures, including QA/QC, are vital to the reliability of the program. There is no doubt that it will take significant effort on NCD’s part to memorialize its policies and procedures, but without such policies and procedures and the consistent application of them, stakeholders will lack confidence that EPA is implementing TSCA in accordance with Congress’s intent.

We are pleased that the allegations that management pressured and incentivized findings of no risk were determined to be without merit. It was not our experience that NCD assessors were neglecting risks. In fact, the rate that PMNs have faced regulations in the form of orders and/or SNURs belies that inference. About 84 percent of PMN determinations have resulted in regulations or have been withdrawn by submitters.