GlobalChem: The search for a fix to IRIS assessments

The EPA is moving ahead with reforms to its Integrated Risk Information System (IRIS) programme, but industry and Congress insist that more needs to be done

The chemical assessments conducted by the US Environmental Protection Agency (EPA) under the Integrated Risk Information System (IRIS) programme are relied upon broadly to provide information on potential health effects from chemical exposure. They underline many federal and state regulatory actions The EPA is reforming the programme and has already begun regular public meetings with stakeholders in a bid to improve transparency.

Copyright: Alamy

IRIS provides information about the health effects associated with chemicals to which people may be exposed over their lifetime. The EPA and US states use IRIS as a primary source of information for environmental health decision-making and its database now contains more than 500 chemicals, but the majority of the files are more than 10 years old.

In the past two years, the EPA has completed seven assessments. Currently, there are nearly 50 substances under consideration for review by the IRIS programme, according to the IRIS assessment tracking system. Over a dozen of those chemicals have schedules in place. The agency is “actively working” on those assessments and plans to add “additional schedules in the near future”.

This year, the EPA plans to send draft assessments for ethylene oxide (EO) and benzo[a]pyrene to the Chemical Assessment Advisory Committee for review. This will also carry out an external peer review of the draft assessments for ammonia and trimethylbenzenes.

CHANGE IS AFOOT
After being heavily criticised by the National Research Council (NRC) in 2011 for its draft formaldehyde assessment, IRIS assessments and processes are undergoing major changes. In its formaldehyde report, the NRC said that EPA’s draft assessment was “not prepared in a logically consistent fashion, lacks clear links to an underlying conceptual framework and does not sufficiently document methods and criteria used to identify evidence for selecting and evaluating studies”.

The NRC made several recommendations for improving the development not only of the formaldehyde assessment but also other IRIS assessments. The NRC also devoted an entire chapter to the overall IRIS assessment process, recommending significant improvements to the methods for gathering, analysing and integrating data on health effects. And, in a 2012 report to Congress, the EPA said it was using a “phased approach” to implement the recommendations.

The EPA is making changes to improve the scientific integrity of assessments and enhance the productivity of the programme. It is also working on increasing transparency so that key scientific issues are identified and debated early in the process.

Both industry and environmental groups have welcomed the proposed changes. The American Chemistry Council’s (ACC) senior director Nancy Beck concedes there has been “transparency on a lot of different levels” but says what has been done so far has been the “easy stuff”. The agency has a “long way to go in transparency regarding criteria for study quality, criteria for weight-of-evidence integration and criteria used to judge quality of specific studies”, she adds.

One of the areas where there has been improvement is the programme’s involvement of stakeholders, says ACC’s senior toxicologist Rick Becker. Central to this are a series of bi-monthly stakeholder meetings, the first of which was held in December 2013. Although the meeting was a “good start,” it was clear that the EPA needs to give more information to stakeholders in advance, says Becker. He suggests that the EPA should identify key risk assessment questions and approaches before the meetings to allow robust scientific discussions about the methods and approaches to be taken in risk assessment procedures.

The EPA is already responding positively to stakeholder suggestions made at the December meeting. Henceforth for such meetings, “Step 1” material will include information about why the EPA is undertaking an assessment. The evidence table may also have data about study methods and quality that will be “important in subsequent assessment development,” says the agency.

The move, the agency says, will provide “additional transparency and more information about our decision-making process for developing the assessment”. To accommodate the changes, the EPA has postponed a scheduled February meeting until 23-24 April.

“Ultimately, these more comprehensive early materials will allow us to have more robust discussion at the IRIS bimonthly meetings,” explained the agency when announcing the meeting’s postponement.

The Natural Resources Defense Council (NRDC) generally supports the proposed changes to the IRIS process, says its senior scientist Jennifer Sass. But the programme must “stick to its deadlines so that its assessments are no longer subjected to the years-long delays of what I hope is the past,” she adds.

The IRIS staff are developing tools to make the process more transparent, she says. “Increasing transparency about which studies they review and how they make selections will be helpful generally, as long as it doesn’t take up too much resources and divert staff time and resources from the real goal of completing assessments.”

She considers that the bi-monthly IRIS meeting will he helpful so long as the agency arranges for web or dial-in access and “sets the agenda in advance so the public can come prepared”.

SYSTEMATIC REVIEW
As part of IRIS reforms, the agency has announced that the programme will adopt systematic review and weight-of-evidence approaches. However, what is still unclear is the timing of when the public will see these new approaches implemented. Kimberly Wise, a senior director with ACC, would like to see systematic reviews to be carried out in a “clear-cut transparent way”, with the agency revealing how it acquires data and how it plans to integrate them into the assessments to arrive at conclusions. Also important is the need to have “clear criteria” for how the agency evaluates study quality, she adds.

Becker concedes the EPA’s recent actions have led to “a little more transparency” in terms of presenting the data. “They have taken the first steps,” he says. “What they haven’t done is implement consistent protocols or procedures for conducting the data evaluation to establish quality and reliability of each study.”

He also criticises the agency for not setting a “transparent framework for integrating the information using weight-of-evidence for mode of action to determine the potential risks to humans. We are optimistically waiting to see those steps.”

The ACC would like a third-party “honest broker” to ensure that the scientific and public comments that the agency receives from peer reviewers and the public are addressed in the assessments. But the EPA says that it already has many reviewers who are “independent of the IRIS programme,” including scientists from the EPA and other federal agencies, who evaluate the EPA’s response to peer review and public comments.

However, the ACC does not think these types of informal interagency reviews are sufficient. Even during these processes, the EPA staff are judge and jury regarding which interagency comments they address and which they ignore and how they are addressed.

ACC would like to see an independent party play a role similar to the journal editor role, to ensure that the Agency is truly responsive, particularly to scientific concerns.

The agency’s Science Advisory Board’s Chemical Assessment Advisory Committee will conduct reviews of IRIS assessments on an ongoing basis, the EPA says. “If, in the course of responding to the peer review recommendations, we encounter significant new scientific issues, then we will conduct additional peer review.”

“While some important incremental changes have been made, most notably with IRIS under Ken Olden’s leadership, more needs to be done to strengthen the scientific underpinnings of these programmes,” Becker concludes. “Specifically, fundamental improvements still must be made in with problem formulation, data evaluation, and use of mode of action in evidence integration to understand cause and effect and to establish realistic health risk values for environmentally relevant exposure levels.”

Dinesh Kumar is a reporter for Chemical Watch, an online service which provides news and intelligence on global chemical regulations. As part of Chemical Watch’s US-based editorial team, he reports on ­domestic chemical regulatory issues

GMU REVEALS CONCERN OVER RISK ASSESSMENTS
A NEW survey on regulatory risk assessment in the US by George Mason University of Arlington,Virginia, finds widespread concern among scientific experts in toxicology and risk assessment over the application of risk assessment procedures.

According to study director Robert Lichter, “There is a gap between what toxicology and risk assessment experts believe should be done… and what is actually put in place in the regulatory setting.”

He adds that “scientific experts believe much needs to be done to improve the way scientific information is collected, analysed and used for determining risks.”

The report highlights concerns in a number of areas, notably: problem formulation, data acquisition, weight of evidence, peer review and risk management.

Less than half those surveyed said they believed risk management decisions were often or always based on current understanding of biology and toxicology. They believe risk managers give too much weight to political concerns, the media and environmental groups, and tool little weight to scientific factors and economic costs and benefits.

You can see the survey at http://bit.ly/1lCuYrU